Since 2ab97c01 "introduce vy_stmt_env and vinyl_max_tuple_size option"
Vinyl's memory is limited both by `struct quota` and `struct vy_quota`.
This causes sporadic allocation errors because `vy_quota` may
exceed in some cases.

Closes #2629

sysconf(_SC_IOV_MAX) can return -1 on glibc 2.25.90.

Closes #2637

We do not store id = 0 in xlog, yet when parsing the log
file expect space id to be present in a request (a mandatory key).

So a space with id 0 is not possible to recovery from a
snapshot.

Simply reserve id 0 to avoid the problem.

region_truncate() is not required there as box_index_get() cleans up
fiber()->gc anyway - see txn_commit_ro_stmt(). Actually, it's even
dangerous: if fiber_gc() called by box_index_get() happens to free
memory, region_truncate() will crash on the assertion:

  region_truncate: Assertion `region_used(region) >= used' failed.

We will be adding one more key def (so called cmp_def) to index_def
shortly, meanwhile convert index_def::key_def to a pointer,
to ease maintenance of the two key defs..

Refactor dependencies of internal files to avoid linking
libbox by unit tests.

Needed for unit tests #1873.

Avoid dependency on memtx_tuple in SysviewEngine.

Needed for unit tests #1873.

In memtx, key_def is what a user has defined. Follow the same style
in vinyl.

In vinyl, key_def variable name stands for a key definition
extended with the primary key parts. We will begin using the
same technique in memtx, so introduce a consistent naming:
use the word "key_def" for the original key def as defined by
the user, and cmp_def for the key def used to compare tuples,
i.e. extended with primary key parts to produce a stable compare
order.

Begin with introducing the name change in vy_mem.[hc]
Continue on to vy_index, vy_run, vy_range, vy_slice, vy_stmt,
and vy_write/read iterators.

Add missing #include <trivia/config.h>

Fixes #2635

Do not allow ddl if fired from on_replace trigger.

Fixes #1968

Tarantool must not crash if space:format called twice during snapshot.

Fixes #2336

Swallowing errors while reading a vylog file is extremely dangerous,
because vylog files contain metadata layout. Any error in vylog might
result in a data corruption beyond recovery in the future. So we must
fail recovery/snapshot if we fail to read the current vylog file. If
this happens, data can still be restored by reading the previous vylog
file or by recovering vylog using write iterator (the latter is yet to
be implemented).

Rename methods to match the common naming scheme.

Follow up #2619 and #2507

No semantic changes.

Follow up #2619 and #2507

* Add proper checks for premature end of request
* Convert to C

Follow up #2619

Remove function_name, expr, user_name from xrow.c, from
sruct request body_map.

Part of #2619

In order of splitting struct request in more specific structures,
the struct auth_request can be used for authentication.

Create a special request type for call/eval requests.
Struct call_request is much lighter than struct request, and its
decoding is faster for call/eval request.
The new struct call_request allows call/eval without parameters.

In a future we are able to add into struct call_request more
members and merge it with box_function_ctx and lua_function_ctx.

Part of #2619

Inline request_normalize_ops() to vinyl.c.
No semantic changes.

Follow up #2507

In order to create struct call_request it is neccessary to move
call/eval logic into separate call.c/.h files to avoid adding
call_request into more common box.h.

Part of #2619

Fix a bug in restoration LE/LT when last_stmt is not in the mem
(it came from another mem or run).

Fix a bug in restoration after EOF reached.

Add a unit test of mem_iterator_restore.

Fixes #2614

Pass func address via asm immediate arg instead of literal constant.

Follow up #2103

* Pass sync as xrow_header_encode() argument, as suggested by Kostja.
* Add SYNC support to Lua xrow reader.
* Add the test cases.

Follow up #2334

Patch "d130255e" disabled encoding of SYNC in xrow_encode_header()
to avoid writing SYNC value into the disk. Unfortunally, this change
has also broken all other users of xrow_encode_header(), such as
replication (JOIN/SUBSCRIBE requests).

Uncomment row->sync encoding in xrow_header_encode() and
hack xlog_write_row() to avoid writing SYNC into WAL.

Closes #2334

See #2334

Space truncation modifies space cache, so we must serialize it with
alter and other truncations, otherwise an attempt to truncate the same
space concurrently from different fibers will lead to a crash.

Remove libbfd from dependencies and use libunwind for stack traces.

Closes #2103

Closes #2288

Now write iterator see only oldest vlsn. Because of it
the write iterator can not skip or merge or squash
newer statements, even if they are not used by any
transactions. For example, consider one key and its lsns:

      LSNs:  10   20   30   40   50
Read Views:  *         *         *

Statements with lsn 40 can be merged with 50;
20 can be merged with 30.
Old write iterator will merge nothing in that case,
because there is oldest vlsn = 10.

Lets allow the write iterator to see all read views, instead
of only oldest.
Such write iterator can merge multiple lsns subsequences
of each key.
For the example above it will merge in the following way:

      LSNs:   10   20   30   40   50
Read Views:   *         *         *
  Merge and  \__/\_________/\________/
    return:    merge,   merge,   merge
             and return each merged result.

For the considered case the difference is 3 result statements
instead of 5 by the old write iterator.

Closes #1920
Closes #2502
Closes #1824
Closes #2578

Follow up #2507, needed fir unit tests

Closes #2507

Part of #2507

Closes #2543

Disable background and usermode when NOTIFY_SOCKET is set.

Closes #2538