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

*) Additionally added GT iterator
*) Added  test for that

* Added a test that acts as expirationd

This patch doesn't fix the bug itself, but simplifies matters
so that there is no more .inprogress xlog files.

.inprogress xlog files were created to start automatically
from a state when the server crashed with an empty
xlog, and couldn't create a new xlog since a file with the
same LSN in the name already existed. To solve this problem,
simply bump LSN at start if the last file in the recovery
list had zero records in it.

Emtpy xlog files still lead to panic if panic_on_wal_error=true
during recovery.

Add a fix and a test case.
Propagate master error to the replica.
The replica no longer enters into an infinite loop, but stops
with a correct status in this case.

*) corrected light and bps_tree corresponingly

Conflicts:
	doc/sphinx/book/box/box_space.rst

Use a bitmap to represent existing servers.

Fix a typo in panic_on_wal_error initialization for relays.

Fix a bug it reveled: search for the matching vclock
could return no vclock if the first xlog was deleted and
since then we added a new server, and joined immediately
with another replica (init_storage.test.lua details, test case #2)

Conflicts:
	doc/sphinx/book/box/box_index.rst

Conflicts:
	doc/user/stored-procedures.xml

ev_stat already uses inotify. Patch libev to add support for
nanosecond timestams in struct stat.