https://bugs.launchpad.net/tarantool/+bug/1160789

When multiple sessions get disconnected at the same
time, the same fiber is running all disconnect triggers.
This fiber's sid was  not changed between trigger invocations,
and on_disconnect() triggrer of the next disconnected
session would see not its own session id in box.session.id(),
but the previous session's session id.

The fix is to correctly set box.session.id() before handling
any iproto request.
In on_connect() request, sid is 0, but soon it is
reset in session_create().

No test case since the bug requires that there is a full
libev queue, so that multiple on_disconnect() triggers
fire in the same loop.

libreadline already goes with libhistory compiled in,
no reason to add it to the dynamically linked library
list.
Apparently, explicitly listing libhistory leads to
the dl loader pre-loading it instead of the compiled-in
version, which, in turn, leads to a crash on the first
history search in the command line client.

The fix is to remove cmake dependency on libhistory
as well as on libncurses, which we don't need either
(always require libtermcap).

A session trigger would always use the same coroutine, for both connect and
disconnect, as well as for multiple sessions.

If a trigger runs insert() or update(), anything that yields the current
coro, the coro stack would get corrupted for the next parallel invocation.

Use an own coroutine for each invocation of a session trigger.

Tarantool showed random error (in tarantool.log) if client
disconnected: SocketError used filehandle instead errno.

Conflicts:
	third_party/README

Preserve the zero-warnings patch.

fiber.name and fiber->gc_pool.name are always the same thing.
Don't store the name twice.

Reorganize iproto parser (pickle) around two pointers at the request
parsing position and request end.
This makes the code more generic and easily usable in lua and
with new buffered I/O.

This also reduces the current widespread use of tbuf throughout
entire code base (good for future maintainability of the code).

Fix race between fiber_yield_timeout() and completed
request during yield itself. In that case eio_poll will
complete request and free its handle. eio_cancel try to
cancel completed and freed request.
Also, fix unpredictable result during timedout test.

When server wrapper script failed to start
the server, it used to try to send a email
to somebody about this. This has never been
tested, nor used.

At least make sure it works when it does try
to do it.

twistd by default starts with umask 077.
This is why buildbot has umask setting.

rpm by default takes file and directory umask
from the build directory.

When building tarantool rpm under buildbot,
twistd default umask was inherited. Thus
all paths were created in rwx-- mode.

Now add to this cmake bug CMake bug 0012864.
That is, by default, cmake cpack generator
adds %dir directive for every intermediate
dir into the spec.

E.g. if there is a destination file /etc/rc.d/foo,
the spec gets
%dir /etc
%dir /etc/rc.d

Together with the default dir access mask of
twistd this nicely messed up our rpm install: an
install of Tarantool 1.4.8-59 rpm for CentOS would change
/etc and /usr access mode to 700.

Two things are done about that: Tarantool no longer
uses a cpack generated cpeck file. Extraneous %dir
directives are removed from the spec file. This
should minimize the damage of a broken twistd config,
should it happen again.

All buildslaves are reconfigured and restarted with umask 022.

A just-connected session would get stuck not reading input
if only a part of request was received.

-H add headers to the raw format output (xlog and snapshot)
-I read xlog content from the stdin