gh-155941: Close the transport when a plain client_connected_cb raises

[danielKim614]

When a plain-function client_connected_cb passed to asyncio.start_server() raises, StreamReaderProtocol.connection_made() now reports the error via the loop exception handler and closes the transport, mirroring what gh-111601 (gh-110894) did for coroutine callbacks.

Previously the exception propagated out of connection_made(), so the server-side transport stayed open forever, leaking the connection and making Server.wait_closed() hang. The error was also reported with the generic callback message instead of "Unhandled exception in client_connected_cb".

The new test fails without the fix and passes with it. The full test_asyncio suite passes, including -R 3:3 refleak runs on test_streams.

cc. Pycon 2026 KR Sprint @hugovk @corona10

• Issue: asyncio: a plain-function client_connected_cb that raises leaves the server transport open #155941

[danielKim614]

@kumaraditya303 Should this be backported to 3.13 and 3.14?
The bug is present there as well, and the leaked connection also makes Server.wait_closed() hang.
Happy to resolve the conflicts manually if the bot backports fail.

[miss-islington-app]

Thanks @danielKim614 for the PR, and @kumaraditya303 for merging it 🌮🎉.. I'm working now to backport this PR to: 3.14.
🐍🍒⛏🤖

[miss-islington-app]

Thanks @danielKim614 for the PR, and @kumaraditya303 for merging it 🌮🎉.. I'm working now to backport this PR to: 3.15.
🐍🍒⛏🤖 I'm not a witch! I'm not a witch!

[bedevere-app]

GH-156273 is a backport of this pull request to the 3.14 branch.

[bedevere-app]

GH-156274 is a backport of this pull request to the 3.15 branch.