…2 bug that swaps DAC channels randomly.

- Fixes adafruit#10003.

Stumbled on espressif/esp-idf#11425 bug while testing. It
causes the left and right channels to be swapped randomly on play, which was very confusing.
Worked around by forcing a deinit/init on each play.

Also got confused because board.A0 and board.A1 are not consistently assigned to left/right
channels. Added a documentation warning about this.

Fix crash on deinit of incompletely initied state machine.
Fix processor-dependent includes.

This test was written in such a way that having a wrong data type
for the mp3 samples wasn't detected. Instead of using
np.frombuffer(dtype=int16), just do arithmetic directly on the
samples. During testing time we don't care if it might be a little
slower or use a little more RAM than ulab, and we don't care
whether it's actually an RMS calculation.  Just that it's
consistent and shows the audio data is correct, including its
defined data type.

This fixes an unlikely problem with the USB host implementation on
rp2350 that would not have detected failure to allocate a DMA channel.

Together with adafruit#10186 this should give a clean build. As it is,
this will error.

mp3 samples are signed

Disallow identical AudioOut channel pins. Work around ESP-IDF ESP32-S2 bug that swaps DAC channels randomly.

Fix signedness bug causing PIO hang-up.

This lets each MCU type get the correct definitions. It also simplifies
paths at include sites.

Closes: adafruit#10181

Enable & Address -Wtype-limits diagnostics in raspberrypi port

Depend on header inclusion via -isystem

…le release of released resources.

Add guards to common_hal_pulseio_pulse(in|out)_deinit to prevent doub…

…_92x

fix tilegrid.contains for subclasses