This applies only to duplex devices.

This unifies, as much as possible, the ScriptProcessorNode path with
the AudioWorklets path to avoid some code duplication, and to also make
the two paths more similar to each other to ease in maintenance.

  * Duplex mode now only creates a single AudioContext and AudioWorklet
  * Devices should now automatically start in response to a gesture

Public issue https://github.com/mackron/miniaudio/issues/597

This adds the following APIs:

  * ma_sound_stop_with_fade_in_pcm_frames()
  * ma_sound_stop_with_fade_in_milliseconds()
  * ma_sound_set_stop_time_with_fade_in_pcm_frames()
  * ma_sound_set_stop_time_with_fade_in_milliseconds()

These functions will overwrite any existing fades. For the
set_stop_time variants, you specify the time that the sound will be put
into it's stopped state. The fade will start at stopTime - fadeLength.
If the fade length is greater than the stop time, the fade length will
be clamped.

Public issue https://github.com/mackron/miniaudio/issues/669

This adds the following APIs:

  * ma_sound_set_fade_start_in_pcm_frames()
  * ma_sound_set_fade_start_in_milliseconds()

Public issue https://github.com/mackron/miniaudio/issues/669

The sample format of a WAV file is not always being set which results
in get_data_format() returning ma_format_unknown.

This change makes use of the onInitFile, onInitFileW and onInitMemory
backend callbacks which enables decoding backends to have optimized
implementations for reading from a file or a block of memory without
having to go through an abstraction layer on the miniaudio side.

Public issue https://github.com/mackron/miniaudio/issues/696

Public issue https://github.com/mackron/miniaudio/issues/707

Public issue https://github.com/mackron/miniaudio/issues/700

When this is used, the data callback should at some point call
ma_engine_read_pcm_frames() in order to do some processing.

This is optional and is fired at the end of each call to
ma_engine_read_process_pcm_frames(). The callback will be passed the
processed audio data so they can do their own processing such as
outputting to a file or whatnot.

The callback is configured via the engine config.