gh-87135: Raise PythonFinalizationError when joining a blocked daemon thread (gh-130402)

If `Py_IsFinalizing()` is true, non-daemon threads (other than the current one)
are done, and daemon threads are prevented from running, so they
cannot finalize themselves and become done. Joining them (without timeout)
would block forever.

Raise PythonFinalizationError instead of hanging.

Raise even when a timeout is given, for consistency with trying to join your own thread.

See gh-123940 for a use case: calling `join()` from `__del__`. This is
ill-advised, but an exception should at least make it easier to diagnose.

Lib/test/test_threading.py

@@ -1171,6 +1171,77 @@ def __del__(self):
         self.assertEqual(out.strip(), b"OK")
         self.assertIn(b"can't create new thread at interpreter shutdown", err)
 
+    def test_join_daemon_thread_in_finalization(self):
+        # gh-123940: Py_Finalize() prevents other threads from running Python
+        # code, so join() can not succeed unless the thread is already done.
+        # (Non-Python threads, that is `threading._DummyThread`, can't be
+        # joined at all.)
+        # We raise an exception rather than hang.
+        for timeout in (None, 10):
+            with self.subTest(timeout=timeout):
+                code = textwrap.dedent(f"""
+                    import threading
+
+
+                    def loop():
+                        while True:
+                            pass
+
+
+                    class Cycle:
+                        def __init__(self):
+                            self.self_ref = self
+                            self.thr = threading.Thread(
+                                target=loop, daemon=True)
+                            self.thr.start()
+
+                        def __del__(self):
+                            assert self.thr.is_alive()
+                            try:
+                                self.thr.join(timeout={timeout})
+                            except PythonFinalizationError:
+                                assert self.thr.is_alive()
+                                print('got the correct exception!')
+
+                    # Cycle holds a reference to itself, which ensures it is
+                    # cleaned up during the GC that runs after daemon threads
+                    # have been forced to exit during finalization.
+                    Cycle()
+                """)
+                rc, out, err = assert_python_ok("-c", code)
+                self.assertEqual(err, b"")
+                self.assertIn(b"got the correct exception", out)
+
+    def test_join_finished_daemon_thread_in_finalization(self):
+        # (see previous test)
+        # If the thread is already finished, join() succeeds.
+        code = textwrap.dedent("""
+            import threading
+            done = threading.Event()
+
+            def loop():
+                done.set()
+
+
+            class Cycle:
+                def __init__(self):
+                    self.self_ref = self
+                    self.thr = threading.Thread(target=loop, daemon=True)
+                    self.thr.start()
+                    done.wait()
+
+                def __del__(self):
+                    assert not self.thr.is_alive()
+                    self.thr.join()
+                    assert not self.thr.is_alive()
+                    print('all clear!')
+
+            Cycle()
+        """)
+        rc, out, err = assert_python_ok("-c", code)
+        self.assertEqual(err, b"")
+        self.assertIn(b"all clear", out)
+
     def test_start_new_thread_failed(self):
         # gh-109746: if Python fails to start newly created thread
         # due to failure of underlying PyThread_start_new_thread() call,