This allows us to properly update the ALCdevice and its resources with the new
parameters before starting playback, instead of expecting the mixer to block
and wait after it has begun.
This also lets us avoid holding the device lock while resetting and starting
the device, which helps prevent lock inversion on some backends (ie, one thread
locking A then B, and another thread locking B then A), ultimately allowing
certain backends to asynchronously update the ALCdevice without risk of lockup.
Capture still has issues here, however.
The processing thread must be messaged when shutting down, and depending on the
state of the device, waveInReset may not do it. Additionally, all buffers must
be removed from the device before they can be unprepared and the device closed.
If the processing threads are interrupted between the shutdown check and re-
adding the buffer back to the device, the device can be left with a buffer that
will get unprepared and freed while in use.
