applet.memory.prom: modify to use port groups.

[q3k]

Part of #599.

This one fails with a nasty stacktrace if --pins-a isn't set:

$ venv/bin/glasgow build  --rev C3 memory-prom
Traceback (most recent call last):
  File "/home/q3k/Projects/glasgow2/software/venv/bin/glasgow", line 8, in <module>
    sys.exit(run_main())
             ^^^^^^^^^^
  File "/home/q3k/Projects/glasgow2/software/glasgow/cli.py", line 939, in run_main
    exit(asyncio.run(main()))
         ^^^^^^^^^^^^^^^^^^^
  File "/run/current-system/sw/lib/python3.11/asyncio/runners.py", line 190, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/run/current-system/sw/lib/python3.11/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/run/current-system/sw/lib/python3.11/asyncio/base_events.py", line 654, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/q3k/Projects/glasgow2/software/glasgow/cli.py", line 822, in main
    target, applet = _applet(args.rev, args)
                     ^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/q3k/Projects/glasgow2/software/glasgow/cli.py", line 424, in _applet
    applet.build(target, args)
  File "/home/q3k/Projects/glasgow2/software/glasgow/applet/memory/prom/__init__.py", line 522, in build
    ports=iface.get_port_group(
          ^^^^^^^^^^^^^^^^^^^^^
  File "/home/q3k/Projects/glasgow2/software/glasgow/access/__init__.py", line 73, in get_port_group
    return PortGroup(**{
                       ^
  File "/home/q3k/Projects/glasgow2/software/glasgow/access/__init__.py", line 74, in <dictcomp>
    name: self.get_port(pin_or_pins, name=name) for name, pin_or_pins in kwargs.items()
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/q3k/Projects/glasgow2/software/glasgow/access/direct/multiplexer.py", line 226, in get_port
    assert port is not None

Not sure what to do about it. I guess not having address pins is an error that was just not caught so far?

[whitequark]

That could be a bug, I'll look.

[whitequark]

Not sure what to do about it. I guess not having address pins is an error that was just not caught so far?

OK so, this assertion appears if you have an empty list of pins rather than None. I didn't know whether this situation will ever arise and I guess now I do :)

[whitequark]

@q3k The nasty backtrace is going to be fixed in #609. Since using zero address pins are a valid configuraiton for this applet (in case you want only the shift register, to save pins), please check that it works.

[whitequark]

@q3k Do you recall what's needed to finish this?