Do not use asyncio in weakref.finalize

[tfarago]

because the ThreadPoolExecutor of asyncio streams complains about submitting things after interpreter shutdown. Instead of trying to be super-clever, I just use the standard sockets, which work and there is no magic. @MarcusZuber this is one of the reasons why async destructors would bite us in our asses in the future.

Traceback (most recent call last):
  File "/usr/lib/python3.12/weakref.py", line 666, in _exitfunc
    f()
  File "/usr/lib/python3.12/weakref.py", line 590, in __call__
    return info.func(*info.args, **(info.kwargs or {}))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/tomas/.local/lib/python3.12/site-packages/concert/devices/cameras/uca.py", line 306, in _ucad_unregister_all
    asyncio.run(_ucad_communicate(struct.pack("I", 14), host, port))
  File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/base_events.py", line 687, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/tomas/.local/lib/python3.12/site-packages/concert/devices/cameras/uca.py", line 295, in _ucad_communicate
    reader, writer = await asyncio.open_connection(host, port)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/streams.py", line 48, in open_connection
    transport, _ = await loop.create_connection(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/base_events.py", line 1080, in create_connection
    infos = await self._ensure_resolved(
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/base_events.py", line 1456, in _ensure_resolved
    return await loop.getaddrinfo(host, port, family=family, type=type,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/base_events.py", line 901, in getaddrinfo
    return await self.run_in_executor(
                 ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/base_events.py", line 863, in run_in_executor
    executor.submit(func, *args), loop=self)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/concurrent/futures/thread.py", line 172, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown

Alternative solution (which would enable custom concert shutdown hook)

Rough code:

$ concert start adestruct
adestruct > from concert.devices.cameras.uca import RemoteNetCamera, _ucad_unregister_all
adestruct > def mine(info):
       ...:     if info.raw_cell == "exit":
       ...:         _ucad_unregister_all("localhost", 8989)
       ...: 
adestruct > ip = get_ipython()
adestruct > ip.events.register("pre_run_cell", mine)
adestruct > a = 1
# Nothing happens
adestruct > exit
# on ucad: ** (ucad:205608): DEBUG: 12:28:02.831: All endpoints removed

Actual simple concert implementation is in this commit.

[tfarago]

If you guys agree I would roll this out ASAP, it causes a page-long error on concert exit.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.52%. Comparing base (64508fd) to head (2c951a3).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #562   +/-   ##
=======================================
  Coverage   83.52%   83.52%           
=======================================
  Files         133      133           
  Lines       10718    10718           
=======================================
  Hits         8952     8952           
  Misses       1766     1766           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[tfarago]

@MarcusZuber the "alternative" solution would enable async destructors, not per se, but by registering some cleanup actions in concert.

[MarcusZuber]

This fixes it here and we could merge it.

Having the possibility to use "proper" destructors would also be nice to have.

[tfarago]

I am a little bit afraid when it comes to the ipython magic but it avoids duplicating async code with a sync version. So I will implement that and in the worst case we change it later.