The test is flaky, because it tries to listen on a fixed TCP port, which
may be busy. Let's use Unix sockets to fix the test.

While we are at it, let's also remove useless creation of a temporary
directory - the test copies a script there, but never uses it.

The test injects an error for a replication of a particular LSN. The
problem there's a pending record about the new replica written to the
_cluster space on join, which may not have been flushed to WAL by the
time the test injects an error. Due to the race, the test may inject an
error for the _cluster row instead of the test row, which breaks its
expectations.

```
[009] xlog/panic_on_broken_lsn.test.lua                               [ fail ]
[009]
[009] Test failed! Result content mismatch:
[009] --- xlog/panic_on_broken_lsn.result       Tue Nov  2 13:29:53 2021
[009] +++ var/rejects/xlog/panic_on_broken_lsn.reject   Tue Nov  2 13:29:59 2021
[009] @@ -155,8 +155,8 @@
[009]  ...
[009]  (found:gsub('^.*, req: ', ''):gsub('lsn: %d+', 'lsn: <lsn>'))
[009]  ---
[009] -- '{type: ''INSERT'', replica_id: 1, lsn: <lsn>, space_id: 9000, index_id: 0, tuple:
[009] -  [2, "v1"]}'
[009] +- '{type: ''INSERT'', replica_id: 1, lsn: <lsn>, space_id: 320, index_id: 0, tuple:
[009] +  [2, "34898d18-eed4-4f8a-97ff-4ffba7b42892"]}'
[009]  ...
[009]  test_run:cmd('cleanup server replica')
[009]  ---
[009]
```

Fix this by flushing WAL before writing a broken row.

Closes #4508

This patch extends the 'integration.yml' workflow and adds a new
workflow call for running tests to verify integration between tarantool
and the 'checks' module.

Part of #5265
Part of #6056
Closes #6562

Part of #6257

@TarantoolBot document
Title: Document net.box watchers

Using the new `watch` method of a net.box connection, one can subscribe
to events broadcasted by a remote host. The method has the same syntax
as the `box.watch()` function, which is used for subscribing to events
locally. It takes a key name (string) to subscribe to and a callback to
invoke when the key value is updated. It returns a watcher handle that
can be used to unregister the watcher. Note, garbage collection of a
watcher handle doesnt result in unregistering the watcher so it's okay
to discard the result of `box.watch` if the watcher is never going to be
unregistered.

A watcher callback is first invoked unconditionally after the watcher
registration. Subsequent invocations are triggered by `box.broadcast()`
called on the remote host. A watcher callback is passed the name of the
key the watcher was subscribed to and the current key value. A watcher
callback is always executed in a new fiber so it's okay to yield inside
it. A watcher callback never runs in parallel with itself: if the key
to which a watcher is subscribed is updated while the watcher callback
is running, the callback will be invoked again with the new value as
soon as it returns.

Watchers survive reconnect (see `reconnect_after` connection option):
all registered watchers are automatically resubscribed as soon as the
connection is reestablished.

If a remote host supports watchers, the 'watchers' key will be set in
connection's `peer_protocol_features`.

Example usage:

 * Server:
   ```lua
   -- Broadcast value 123 for key 'foo'.
   box.broadcast('foo', 123)
   ```
 * Client:
   ```lua
   conn = net.box.connect(URI)
   -- Subscribe to updates of key 'foo'.
   w = conn:watch('foo', function(key, value)
       assert(key == 'foo')
       -- do something with value
   end)
   -- Unregister the watcher when it's no longer needed.
   w:unregister()
   ```

Part of #6257

@TarantoolBot document
Title: Document IPROTO watchers

There are three new commands to support asynchronous server->client
notifications signaled with `box.broadcast()`:

 - `IPROTO_WATCH` (code 74). Registers a new watcher for the given
   notification key or acknowledges a notification if a watcher is
   already registered. The key name is passed in `IPROTO_EVENT_KEY`
   (code 0x56). The watcher will be notified unconditionally after
   registration and then every time the key is updated with
   `box.broadcast()` provided the last notification was acknowledged.
   The server doesn't reply to the request unless it fails to parse
   the packet.

 - `IPROTO_UNWATCH` (code 75). Unregisters a watcher registered for the
   given notification key. The key name is passed in `IPROTO_EVENT_KEY`
   (code 0x56). A server doesn't reply to the request unless it fails to
   parse the packet.

 - `IPROTO_EVENT` (code 76). Sent by the server to notify a client
   about a key update. The key name is passed in `IPROTO_EVENT_KEY`
   (code 0x56). The key data (optional) is passed in `IPROTO_EVENT_DATA`
   (code 0x57).

When a connection is closed, all watchers registered for it are
unregistered.

Servers that support the new feature set the `IPROTO_FEATURE_WATCHERS`
feature bit (bit 3) in reply to the `IPROTO_ID` command.

Part of #6257

@TarantoolBot document
Title: Document box.watch and box.broadcast

`box.watch(key, func)` registers a watcher for the given key and returns
a watcher handle, which can be used to unregister the watcher (by
calling the `unregister` method). A key is an arbitrary string. It's
possible to register more than one watcher for the same key. Note,
garbage collection of a watcher handle doesnt result in unregistering
the watcher so it's okay to discard the result of `box.watch` if the
watcher is never going to be unregistered.

`box.broadcast(key, value)` updates the value of the given key and
signals all watchers registered for it.

A watcher callback is first invoked unconditionally after the watcher
registration. Subsequent invocations are triggered by `box.broadcast()`
called on the local host. A watcher callback is passed the name of the
key the watcher was subscribed to and the current key value. A watcher
callback is always executed in a new fiber so it's okay to yield inside
it. A watcher callback never runs in parallel with itself: if the key
to which a watcher is subscribed is updated while the watcher callback
is running, the callback will be invoked again with the new value as
soon as it returns.

`box.watch` and `box.broadcast` may be used before `box.cfg`.

Example usage:

```lua
-- Broadcast value 123 for key 'foo'.
box.broadcast('foo', 123)
-- Subscribe to updates of key 'foo'.
w = box.watch('foo', function(key, value)
    assert(key == 'foo')
    -- do something with value
end)
-- Unregister the watcher when it's no longer needed.
w:unregister()
```

Part of #6257

This commit introduces a C module that will later be used for
implementation of the watchers API in both Lua and IPROTO.

Key points:
 - A watcher callback can be registered for an arbitrary string key with
   box_register_watcher().
 - Watcher callbacks are invoked in the background by the worker fiber.
   If a watcher is asynchronous (WATCHER_RUN_ASYNC flag set), the worker
   fiber will invoke the callback in a new fiber.
 - A newly registered watcher callback is scheduled for execution
   unconditionally after registration, and then whenever box_broadcast()
   is called for the specified key.
 - The caller may pass arbitrary data to box_broadcast() which will be
   stored internally and passed to registered callbacks.
 - A callback is not executed again until it acknowledges the last
   notification (explicitly by calling watcher_ack() or implicitly by
   returning if WATCHER_EXPLICIT_ACK is unset).

To be honest, I have not proceeded with necessary actions to receive
personal Telegram notifications for my branches, because GitHub already
sends them to my email.

I kept this feature (I don't know, whether someone uses it), but let the
action ignore failures in the case.

A failed attempt to send the notification from `master`, a release
branch (`1.10`, `2.8`, ...) or a tag will fail the job as before. The
change affects only personal notifications for developer branches.

Fixes https://github.com/tarantool/tarantool-qa/issues/130

There are two kind of schema objects: objects themselves (featuring
specific ids) and entities (representing all kinds of objects of the
concrete type). Granting privileges for entities from user different from
admin is not allowed. In this case diag is raised. Accidentally, there
was a bug in diag: instead of printing entity name attempt to print
object name took place. That led to assertion fault on debug version.
Let's fix this and print entity name as it supposed to be.

Closes #5389

enum schema_object_type fits values of two different ranges, so let's
add converters to transform from one to another.

We have limited amount of arm64 runners and arm64 jobs are often queued
for hours. So I disabled running of arm64 jobs on developer branches and
external PRs. Now we run them on master, release branches, tags and for
branches named as 'something-full-ci'.

While I'm here, disabled those jobs explicitly on forks: I guess
otherwise it'll be it a 'waiting a runner' state infinitely.

Part of https://github.com/tarantool/tarantool-qa/issues/142

After testing is enabled for aarch64 deployment,
it was revealed that we lack following packages for
CentOS 7 / aarch64:
       python-gevent3, python-pyyaml

After we'll backport the packages to packpack
deployment can be enabled again.

Except for backtrace related stuff, see gh-6060 for details.

This partially reverts commit 24c065d6.

Also rename suffices for ARM64 to aarch64 in
corresponding workflows.

Also, remove explicit RelWithDebInfo teting  workflow
for aarch64 since it is now tested by packeting workflows.

This patch extends the 'integration.yml' workflow and adds a new
workflow call for running tests to verify integration between tarantool
and the queue module.

Part of #5265
Part of #6056
Closes #6559

Replication reconfiguration used to work as follows: upon receiving a
new config disconnect from all the existing masters and try to connect
to all the masters from the new config.

This lead to instances doing extra work when old config and new config
had the same nodes in them: instead of doing nothing, tarantool
reinitialized the connection.

There was another problem: when an existing connection is broken, master
takes some time to notice that. So, when replica resets the connection,
it may try to reconnect faster than the master is able to notice its
absence. In this case replica wouldn't be able to reconnect due to
`duplicate connection with the same replica UUID`. So replication would
stop for a replication_timeout, which may be quite large (seconds or
tens of seconds).

Let's prevent tarantool from reconnecting to the same instance, if there
already is a working connection.

Closes #4669

Once an anonymous replica tries to register, it reconnects to every
other instance in its replication list in order to receive updated
ballots and choose someone to register on.

Make the instance wait until it connects to quorum before letting it
choose the node to register on.

Since the function arguments are always sequential in the array of all
VDBE MEMs, we don't need to store the position of each argument. The
position of the first MEM and the number of arguments are sufficient to
describe all the arguments.

Part of #4145

This patch removed MEM_TYPE_AGG, which is no longer used due to recent
changes.

Part of #4145

This patch removes copying of the result in the finalizers of the SQL
built-in aggregate functions.

Part of #4145

Part of #4145

Part of #4145

Part of #4145

Part of #4145

Part of #4145

Part of #4145

We don't need this function, since it is easier to call finalizers
directly. This patch also allows us to make further simplifications.

Needed for #4145

This patch introduces the mem_append() function. This function appends
the specified string to the end of the STRING or VARBINARY contained in
MEM. In case MEM needs to increase the size of allocated memory,
extra memory is allocated in an attempt to reduce the total number of
allocations.

Needed for #4145

This patch makes it easier to get a collation by a function.

Needed for #4145

This patch moves the initialization of sql_context out of the VDBE. This
allows us to remove the opcodes OP_BuiltinFunction0 and OP_AggStep0,
which work in a rather strange way. Moreover, due to the changes these
opcodes make to the VDBEs, it is possible that the estimated size of the
VDBE may change during execution of VDBE, which could lead to various
problems.

Needed for #4145

This patch makes OP_FunctionByName, OP_AggStep and OP_BuiltinFunction to
use register P1 for the number of arguments instead of register P5. This
makes it easier to use these opcodes.

Needed for #4145

This patch fixes possible undefined behavior during the implicit cast of
INTEGER to DOUBLE. The problem is, if the INTEGER is close enough to
2^64, it will be cast to 2^64 when it is cast to DOUBLE. Since we have a
check for loss of precision, this will cause this DOUBLE to be cast to
an INTEGER, which will result in undefined behavior since this DOUBLE is
outside the range of INTEGER.

This patch removes zeroblob optimization from SQL code. This
optimization complicates the code, and there is almost no profit from
it.

Closes #6113
Needed for #4145

This patch extends the 'integration.yml' workflow and adds a new
workflow call for running tests to verify integration between tarantool
and the avro-schema module.

Part of #5265
Part of #6056
Closes #6558

This patch extends the 'integration.yml' workflow and adds a new
workflow call for running tests to verify integration between tarantool
and the metrics module.

Part of tarantool/tarantool#5265
Part of tarantool/tarantool#6056
Closes tarantool/tarantool#6526

*.inprogress files shouldn't be deleted during normal GC, because they
may be in use. E.g. GC may happen to run while WAL is rotating xlog;
if GC removes the transient xlog.inprogress file created by WAL (see
xdir_create_xlog), WAL will fail to rename it and abort the current
transaction.

Initially, inprogress files were removed only after recovery. For this
reason, xdir_collect_inprogress, which is used for deleting inprogress
files, accesses FS directly, without COIO. This was changed by commit
5aa243de ("recovery: build secondary
index in hot standby mode") which moved xdir_collect_inprogress from
memtx_engine_end_recovery to memtx_engine_collect_garbage so that a hot
standby instance doesn't delete inprogress files of the master instance
by mistake.

To fix this issue, let's move xdir_collect_inprogress back where it
belongs, to engine_end_recovery, and introduce a new callback for memtx
to build its secondary keys before entering the hot standby mode -
engine_begin_hot_standby.

Let's also remove engine_collect_garbage from gc_init, which was added
there by the aforementioned commit, probably to make tests pass.

The bug was reported by the vinyl/deferred_delete test (#5089).

Closes #6554

Quorum.test.lua was migrated to luatest.
It was divided into several tests.

Delete quorum.test.lua, fix suite.ini in replication tests.

Part of tarantool/test-run#304

Use luatest for checking correctness behavioral for encode/decode
functions

Delete msgpack.test.lua.

Part of tarantool/test-run#304

The commit includes a few helpers files for luatest infrastructure.
'Server' based on luatest Server with additional functionality like
waiting Server for start, create Server's own env including box.cfg
provided by test.

Servers could be built and added in a cluster (cluster.lua file).

Asserts.lua is additional checks, which are useful for many tests.

Helpers are supportive function, could be also used for many tests.

Part of tarantool/test-run#304

Same as for local transactions, timeout for iproto transactions
was implemented. If timeout is not specified in client request
it's sets to box.cfg.txn_timeout, which specified on server side.

Closes #6177

@TarantoolBot document
Title: ability to set timeout for iproto transactions was implemented
A new `IPROTO_TIMEOUT 0x56` key has been added. Currently it is used to
set a timeout for transactions over iproto streams. It is stored in the
body of 'IPROTO_BEGIN' request. If user want's to specify timeout using
netbox (3s for example), he should use 'stream:begin({timeout = 3}).