snd_pcm_writei() is meant to block block until all samples are written,
but apparently in some situations it can block for much longer, prehaps
even a infinite time, in the case of virtual machine FifoCI runs in.

Because it grabed a mutex before blocking, it could also block the
Clear() call for an infinite length of time, blocking dolphin's emu
thread.

snd_pcm_writei() also takes 10-15 seconds if you run dolphin under GDB
and can randomly take 5 or so seconds during normal usage.

By moving all the pause code to the ALSA thread, Clear() no-longer
blocks and everyone keeps their sanity.

When the emulation is paused and the ALSA backend is used, make the audio
thread wait on a condition variable instead of busy-waiting. This commit
fixes bug #7729

Since the ALSA API is not thread-safe, calls to snd_pcm_drop() and snd_pcm_prepare()
in AlsaSound::Clear() are protected by the same mutex as the condition variable in AlsaSound::SoundLoop()
to make sure that we do not call these functions while a call to
snd_pcm_writei() is ongoing.

This fixes a race condition:

Before this commit, there was a race condition when starting a game:

Core::EmuThread(), after having started (but not necessarily completed)
the initialization of the audio thread, calls Core::SetState() which calls
CCPU::EnableStepping(), which in turns calls AudioCommon::ClearAudioBuffer().

This means that SoundStream::Clear() can be called before
AlsaSound::AlsaInit() has completed.

Instead of creating the mixer externally and then passing it in, it can just be made within the class.