This saves about 60 bytes (Feather M4 went from 45040 -> 45100 bytes free)
66 bytes of data eliminated, but 6 bytes paid back to initialize the length
field.
These changes remove the caveat from supervisor.runtime.serial_connected.
It appears that _tud_cdc_connected() only tracks explicit changes to the
"DTR" bit, which leads to disconnects not being registered.
Instead:
* when line state is changed explicitly, track the dtr value in
_serial_connected
* when the USB bus is suspended, set _serial_connected to False
Testing performed (using sam e54 xplained): Run a program to show
the state of `serial_connected` on the LED:
```
import digitalio
import supervisor
import board
led = digitalio.DigitalInOut(board.LED)
while True:
led.switch_to_output(not supervisor.runtime.serial_connected)
```
Try all the following:
* open, close serial terminal program
- LED status tracks whether terminal is open
* turn on/off data lines using the switchable charge-only cable
- LED turns off when switch is in "charger" position
- LED turns back on when switch is in Data position and terminal is
opened (but doesn't turn back on just because switch position is
changed)
This started while adding USB MIDI support (and descriptor support is
in this change.) When seeing that I'd have to implement the MIDI class
logic twice, once for atmel-samd and once for nrf, I decided to refactor
the USB stack so its shared across ports. This has led to a number of
changes that remove items from the ports folder and move them into
supervisor.
Furthermore, we had external SPI flash support for nrf pending so I
factored out the connection between the usb stack and the flash API as
well. This PR also includes the QSPI support for nRF.
