fix: Cleanup prometheus metrics for dead workers

The current "cleanup" function that is removing orphan prometheus files
is actually never invoked.
We move this to a default gunicorn config file to register our hook(s).

https://docs.gunicorn.org/en/stable/configure.html
https://docs.gunicorn.org/en/stable/settings.html#child-exit
Signed-off-by: moson <moson@archlinux.org>

gunicorn.conf.py

@@ -0,0 +1,7 @@
+from prometheus_client import multiprocess
+
+
+def child_exit(server, worker):  # pragma: no cover
+    """This function is required for gunicorn customization
+    of prometheus multiprocessing."""
+    multiprocess.mark_process_dead(worker.pid)