Inspired by issue24391 and the changes proposed to the threading module's reprs for Event, Semaphore, BoundedSemaphore, and Barrier, the corresponding objects in the multiprocessing module should have their reprs updated accordingly.

This patch implements the majority opinion from issue24391 for the desired format of the reprs produced by Event, Semaphore, BoundedSemaphore, and Barrier.  It provides tests around each, inspired by Serhiy's preliminary patch for that same issue but adapted for multiprocessing.

These tests pass on all but 2 -- it is expected that those 2 will pass once the threading module has been similarly updated as some of the combinatoric tests leverage threading.Event in some cases and multiprocessing.Event in others (and so on for the other synchronization objects).  At present, Serhiy's patch still needs updating to match the majority opinion on the repr format, hence it could not yet be applied in combination with this patch for testing.  Tests thus far have been performed on OS X 10.10.3.

To Antoine's point in issue24391, here too, these modifications to the format of the repr are estimated to have very little impact or risk to breaking existing code.

The patch need to be converted to a GitHub PR.

Given https://github.com/python/cpython/pull/20534 is merged and the multiprocessing.synchronize objects inherit from the corresponding threading module objects, is this still needed?

Nihir: in synchronize.py, only Barrier inherits from threading.Barrier, as far as I can see. Other classes inherit from `object`.

New changeset af6b4068859a5d0c8afd696f3c0c0155660211a4 by Kumar Aditya in branch 'main':
bpo-25066: Added repr for multiprocessing.Event (GH-29749)
https://github.com/python/cpython/commit/af6b4068859a5d0c8afd696f3c0c0155660211a4