fix: set a max jobs on win32 to avoid too many handles

This commit fixes the issue of too many handles on win32 when `--jobs=auto`.
This is done by setting the `job_count` to the minimum value between
`multiprocessing.cpu_count()` and 61.  This happens when a machine has >
61 logic processors and `multiprocessing.cpu_count()` returns a value
greater than 61.

See [Python Issue 26903](https://bugs.python.org/issue26903) for more information.

Example failure for Python 3.10.15 with flake8 7.1.0 on Windows 11:

```bash
flake8.EXE myfile.py --jobs=auto
Exception in thread Thread-1 (_handle_workers):
Traceback (most recent call last):
  File "lib\threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "lib\threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "lib\multiprocessing\pool.py", line 522, in _handle_workers
    cls._wait_for_updates(current_sentinels, change_notifier)
  File "lib\multiprocessing\pool.py", line 502, in _wait_for_updates
    wait(sentinels, timeout=timeout)
  File "lib\multiprocessing\connection.py", line 879, in wait
    ready_handles = _exhaustive_wait(waithandle_to_obj.keys(), timeout)
  File "lib\multiprocessing\connection.py", line 811, in _exhaustive_wait
    res = _winapi.WaitForMultipleObjects(L, False, timeout)
ValueError: need at most 63 handles, got a sequence of length 194
```

src/flake8/checker.py

@@ -8,6 +8,7 @@ import logging
 import multiprocessing.pool
 import operator
 import signal
+import sys
 import tokenize
 from typing import Any
 from typing import Generator
@@ -151,7 +152,14 @@ class Manager:
         # default to 1
         if jobs.is_auto:
             try:
-                return multiprocessing.cpu_count()
+                system_cpus = multiprocessing.cpu_count()
+
+                # Work around https://bugs.python.org/issue26903
+                return (
+                    min(system_cpus, 61)
+                    if sys.platform == "win32"
+                    else system_cpus
+                )
             except NotImplementedError:
                 return 0