In the next patches the field_def will being parsed from space:format.
Field_def will contain char *name, which is limited by BOX_NAME_MAX = 65000.
So neither opt_type OPT_STR or OPT_STRPTR can be used to parse this name from
space:format. Besides, field_def contains enum field_type, which can not
be parsed using and opt_type.
Also, field_def will contain default_value, which can store values of
many types.

Proposal is to use opt_create_from_field not for entire field_def, but only
for several fields using opts_parse_key. And parse other options
manualy.

Needed for #2652

replication/cluster.test.py would fail at server exit, because
at_exit() handler tries to destroy a cbus while its mutex is locked.

args.test.py would fail when run with 'make test'

sql/sql-test-name is a strange name for a test.

Implement support for AUTOINCREMENT feature for integer
primary keys of tables which lack rowid property.
All tables which live in Tarantool's universe are lack
rowid.
To implement it two new op-codes were introduced:
  - Fetch maximum integer value of given space/index/fieldno
  - Copy register content from top frame into current frame

Convert AUTOINCREMENT test coverage: sql-tap/autoinc.test

Closes #2263

- Check attached trigger removal when DROP TABLE statement is
executed for referred table
- Implement sql/gh2141-delete-trigger-drop-table.test.lua test
- Assure that #2141 is not a bug, everything works correctly

Closes #2141

This commit also adds some comments which are not important enough to
  have their own commit.

Only 1.7.5 can be upgraded to 1.8 and it is not needed to support
legacy formats.

Now all upgrade_to_1_... can be removed, except 1_8_2(). Inital
function initializes schema for 1.7.5, not for 1.6.0.

We can not support upgrade directly from <1.7.5, because these
versions contains space format conflicts, which are checked in 1.8.

Needed for #2652

* upgrade handler for 1.8.2 must be set to auto instead of
  1.7.5;
* _trigger's first field has type string, but index on this field
  has type unsigned - it is conflict.

Needed for #2652

Implement a special read_string branch for get_next_token function in
sql_tokenizer. Read_string reads entire string, finds single
quote escape and doesn't consider SQL grammar words in the string
as a special words.

- Use indent to convert src/box/sql/pragma.c to linux kernel code style.
- Fix issues with switch indentation
- Update src/lib

Different fibers had shared connection structure - sqlite3 * db.
It had several variables that were shared between transactions in
fibers.

- Move db->autoCommit flag into VDBE
- Move db->isTransactionSavepoint into VDBE
- Move savepoints to VDBE and related counters from struct sqlite3
- Make transactions savepoints local according to VDBE

Closes #2668

fiber_time() reports real time, which shouldn't be used for calculating
timeouts as it is affected by system time changes. Add fiber_clock()
based on ev_monotonic_now(), export it to Lua, and use it instead.

Needed for #2527

We should use ev_monotonic_now()/ev_monotonic_time() instead of
ev_now()/ev_time() for calculating timeouts, because the latter
are affected by system time changes so that using them for timeouts
can lead to unexpected hangs in case system time changes.

Needed for #2527

ev_now()/ev_time() are affected by system time changes, which makes them
impractical for calculating timeouts. Introduce ev_monotonic_now() and
ev_monotonic_time() which work exactly as ev_now()/ev_time(), but use
monotonic clock instead of realtime.

Needed for #2527

Follow up #1265

Follow up #2662

Finally fix compatibility with Lua 5.1 command-line options.

Closes #1265

`tarantool -` now executes stdin instead of trying open "./-" script.

A part of #1265

  -e EXPR - execute string 'EXPR
  -l NAME - require library 'NAME'

A part of #1265

`tarantool -i SCRIPT` forces Tarantool to enter into interactive
mode after executing SCRIPT or stdin.

A part of #1265

getopt_long() is available on all supported platforms.
Get rid of legacy gopt and use getopt_long().

Incompatible changes:

 * `tarantool --version --no-such-option` printed
    "unrecognized option '--no-such-option'", now it displays version.
    `tarantool --no-such-option --version` still prints an error message.

Needed for #1265

- Reference vy_index for every tx value.
- Abort transaction on commit if format was changed by DDL.

Fixes #2342

Check that Tarantool doesn't crash when space remove
while cursor is open.

Fixes #2342

Disconnect replications slave if no ACKs have been received for a
time of `replication_timeout`. This parameter is accessible
via box.cfg{} and applies to all new slave connections.

Follow up #2484

Handle mixed replica_id transaction for wal=none mode.
Follow replicaset vclock after wal writes.
Follow @05eef5fa

On schema version update netbox starts schema reloading but not
disconnects from remote server, so report state as connected for this
case
Fixed #2665

On cpipe_destroy all queued messages were flushed to endpoint but queue
length counter was not zeroed. This breaks an empty pipe check at
cpipe_flush_cb and then crashes tarantool in cases if corresponding endpoint
has been already deleteted before this callback invocation.
Fixed #2703

Replace dynamic function pointers with link-time resolving.

Needed for #2335

Don't resend relay vclock if not changed because a relay loop might be
woken up with cbus messaging.

- In Tarantool there are no limit on space identifier. Check that there
are also no limit on name lengths in SQL.

- Implement sql-tap/gh2127-identifier-max-length.test.lua

Closes #2127

SQLite backup and shell are unnecessary elements, there is no
sense in keeping it in our project.

- Remove src/box/sql/backup.c .
- Remove src/box/sql/shell.c .

Also, fix output for sql/on-conflict

- ON CONFLICT work checked:
Works correctly (not including ROLLBACK, because sql transactions
are not implemented now)
- Tests for ON CONFLICT implemented

Now system space formats contain conflicts between space:format,
their index parts and actual tuple fields.

Such formats can not be used to check tuple fields in SQL.

Fix them in order of #2652