This is a partial fix for an audio regression. The other part of
the fix requires a change in pcaudiolib to reopen the audio device
on flush requests for affected audio drivers/devices.

This reverts commit f9ab812e00.

This fixes a regression introduced in commit
e341068d57.

This is needed for the case when espeak_Cancel is called before
calling espeak_Synth.

This reverts commit 26e453dd2f.

This is based on a patch by Samuel Thibault.

This is needed when espeak_Cancel is called before espeak_Synth,
like can be done in brltty.

This restores the behaviour removed in commit
ca831d236a.

When using eSpeak for audio and eSpeak in a server, using the async
speech requests, that code stopped the audio as soon as possible
from a cancel request. This allows servers like brltty to be more
responsive.

This reverts commit 1b5ea47ea4.

This patch is causing lockups when using this with servers like brltty.

This allows process_espeak_command to stop as soon as possible

Using emacspeak server with espeak-ng (still not published) found that audio was only being stoped after a command is processed (i.e. using pcaudiolib).
This fixes this issue.

Moved include/espeak/speak_lib.h to include/espeak-ng/speak_lib.h and
created a symlink from the original location.

This function should just discard the currently playing audio,
rather than close the audio device.  After all,
we will probably want to synthesize something else later.

Removing these calls do not impact on the behaviour of the program
and simplify the wave.h API.

These calls are used to pre-emptively exit when writing audio
data. This is also handled within dispatch_audio. The difference
is that this callback could allow espeak to cancel quicker. In
practice it does not make much difference.

This call was used to update the written sample count used by
wave_get_remaining_time. As that function has been removed, this
code is no longer needed.