fix(asyncio-cluster): await stale node teardown before replacing cluster nodes

[mkmkme]

Avoid fire-and-forget disconnects when refreshing the async cluster topology.
This keeps removed ClusterNode instances alive until teardown finishes and fixes the unclosed node warning regression during reinitialization.

Fixes #185

Pull Request check-list

• Do tests and lints pass with this change?
• Do the CI tests pass with this change (enable it first in your forked repo and wait for the github action build to finish)?
• Is the new or changed code fully tested?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Is there an example added to the examples folder (if applicable)?

Description of change

[codecov-commenter]

Codecov Report

❌ Patch coverage is 98.94737% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 76.45%. Comparing base (5c9a506) to head (80e7f03).
⚠️ Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
valkey/asyncio/cluster.py	95.45%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #283      +/-   ##
==========================================
+ Coverage   76.41%   76.45%   +0.03%     
==========================================
  Files         129      129              
  Lines       34065    34158      +93     
==========================================
+ Hits        26031    26114      +83     
- Misses       8034     8044      +10     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Pull request overview

This PR addresses an asyncio cluster reinitialization regression by ensuring stale ClusterNode instances are properly disconnected (and awaited) before being dropped/replaced, preventing Unclosed ClusterNode object warnings during topology refresh.

Changes:

• Remove unnecessary asyncio.create_task(...) wrapping around asyncio.gather(...) calls to avoid fire-and-forget behavior.
• Track stale nodes slated for teardown in NodesManager and await their disconnects during initialize() and aclose().
• Add a regression test asserting initialize() waits for removed-node disconnect before completing.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
valkey/asyncio/cluster.py	Introduces pending-node teardown tracking and awaits disconnects during topology refresh and close; simplifies gather usage.
tests/test_asyncio/test_cluster.py	Adds a new test case and helper connection to verify initialization blocks until stale node teardown finishes.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[mkmkme]

@copilot review

[Copilot]

@copilot review

All three review suggestions have been addressed in commit 2ccc88a:

1. Retry on disconnect failure: _close_pending_node_disconnects now uses return_exceptions=True, keeps failed nodes in _pending_node_disconnects for retry on the next call or during aclose(), and only removes entries for successful disconnects.

2. No exception propagation: Disconnect errors emit a RuntimeWarning instead of aborting initialize() or topology refresh. A new test (test_initialize_warns_and_retries_stale_disconnect_errors) covers this path end-to-end, including the retry via aclose().

3. Deterministic test: The test now awaits disconnect_started_event (set at the start of disconnect()) instead of asyncio.sleep(0), ensuring initialize_task is genuinely blocked on the disconnect await before the assertion runs.