Add a directory watcher, in addition to the file watcher.

Chagne wal_dir_rescan_delay default value to 2, since now it's
merely a backup.

When a new log file is created, replication relay may notice
and try to open it before it has a valid header. In this case
the relase is lead into believing that the log file is corrupt.

When opening a new xlog file, first create the file with .inrpogress
suffix, then write the header, thus get a good file, and only then
rename it to a name without a suffix, so that it's noticeable by
the relay. This prevents sporadic failures of replication with
message "failed to read log file header".

The problem was spotted in Avito after a recent upgrade.

Conflicts:
	test/box/sql.result
	test/box/sql.test.lua

Fix a crash in rollbackStatement() when rolling back a
statement trying to access a non-exissting space.

space_cache_find() is called before txn_begin_stmt(), thus
there is no statement object, nothing to roll back.

Remove iditioc tests from the suite.

Change the logic of recovery_finalize() to open
the next xlog if the last xlog had zero rows.
Add a test case.

Remove recovery->finalize, use recovery->writer instead.

Misc. style fixes

Hard-code 1-based error message for now, we'll use field names
in the future anyway.

Remove trailing spaces, fix typos.

Old sql code completly reomoved from tests
box/sql.test, big.sql.test refactoring
sql.tests enabled in big and box suites
Replication tests sql refactoring
parallel call.test sql refactoring
box.error test fixed in call.test

After review fixes:
1. rename remote to iproto
2. remove old sql tests
3. update test-run

*) Splitted rlist initializer to RLIST_HEAD_INITIALIZER(..) and RTREE_LINK_INITIALIZER