Fixes thread sanitizer warnings.

Move its code to DeferEvent instead.

The intrusive contains can easily erase items without searching
through the whole list.  This removes a good amount of runtime
overhead.

By using an "intrusive" data structure, we can easily eliminate
struct TimerRecord.

Prepare to eliminate the TimerRecord struct.

Allows eliminating lots of complexity and workarounds for bogus
assertion failures.

This is only needed while Run() is alive, and moving it there avoids
problems with thread-checking assertions.

Fixes build breakage by commit 4e5271fc; and this method does make
sense in non-debug builds.

Kludge for libavahi-client quirk.

Add the debug-only flag "virgin" which gets checked by assert()
calls.  Fixes assertion failures when using zeroconf/avahi.

Users now have to call Close() explicitly.  This simplifies using the
class, as most users have automatic socket management already, and
Steal() had to be used often.

Add a flag that indicates whether the EventLoop is currently "busy".
As long as that flag is set, it does not need to be woken up - we can
simply add the DeferredMonitor to the list, and it will be caught by
EventLoop very soon.  This eliminates nearly all of the
DeferredMonitor overhead when compared to IdleMonitor, rendering
IdleMonitor mostly obsolete.

Reduce DeferredMonitor overhead.

Improved support for added events.

Now that the remaining known bugs in poll() implementation are fixed,
we can go on without the GLib implementation.

Commit 1f119597 allowed modifying the "quit" attribute from any thread,
and thus the assertion may fail spuriously.  This assertion is too
strict for the relaxed use of "quit".  Let's remove it and move the
"quit" check to before the SockedMonitor::Dispatch() call.

Simply set the "quit" flag and wake up the thread.  This works even if
we're inside this thread.  Setting "quit" to a new value without mutex
protection is usually not safe, but good enough here.

Instead of creating a new eventfd for each DeferredMonitor instance,
reuse EventLoop's eventfd, and add a std::list to EventLoop that
manages the list of pending DeferredMonitors.  This std::list is
protected by the same mutex as the "calls" list.

The bottom line is: reduced overhead because the per-instance eventfd
was eliminated, slightly added overhead due to Mutex usage (but
negligible), and we're thread-safe now.

This subsystem is now good enough to replace EventLoop::AddCall().

This change adds two configuration options:

  --with-eventloop=[glib|internal|auto]
  --with-pollmethod=[epoll|auto]

First allows switching between GLib event loop and internal one.
Second chooses backend to use for internal event loop.
Conditional compilation symbols are changed accordingly.
Additional helper macro MPD_OPTIONAL_FUNC_NODEF is added as well.