fix(weixin): propagate asyncio.TimeoutError from _get_updates to prevent zombie connections

[liuhao1024]

What does this PR do?

Fixes WeChat gateway zombie connection issue where network drops (WiFi reconnect, macOS sleep/wake) cause the adapter to report "connected" indefinitely but no messages can be sent or received.

Root Cause

_get_updates() caught asyncio.TimeoutError and returned an empty success response (ret=0). This caused the poll loop to:

1. Reset consecutive_failures to 0
2. Continue polling on a dead network connection
3. Never detect the connection was lost
4. Report weixin.state=connected forever in /health/detailed

The connection was dead, but the gateway never detected it.

Related Issue

N/A

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)

Changes Made

• See commit messages for detailed changes

How to Test

1. Run pytest tests/ -q — all tests should pass
2. Verify the specific scenario described above is resolved

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: macOS 26.4.1

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture and workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A

[teknium1]

Thanks for isolating the timeout-as-success conversion; current main still has that behavior in gateway/platforms/weixin.py:434-435.

Problems

• Propagation alone does not resolve the reported connected-forever state. The timeout would reach the generic handler at gateway/platforms/weixin.py:1390-1395, which retries and resets the counter but does not call _set_fatal_error, disconnect, or notify the reconnect path. _mark_connected() remains active from gateway/platforms/weixin.py:1295-1296.
• The test hunk is stale: current tests/gateway/test_weixin.py:1194-1207 explicitly asserts the old sentinel behavior. It must be replaced, not supplemented, or the salvaged change leaves contradictory tests.

Suggested changes

• After the selected consecutive-failure threshold, transition through the existing fatal/reconnect lifecycle or rebuild the poll session, and add a poll-loop-level regression for that behavior.
• Update the current sentinel test to require asyncio.TimeoutError propagation.

Automated hermes-sweeper review.

[teknium1]

Propagating the timeout is necessary, but _poll_loop currently catches it generically and only retries/resets its counter; it never clears connected state or enters the fatal/reconnect lifecycle. Please add the threshold transition/rebuild path, otherwise the reported health-state zombie remains.