fix(result-set): bound and clean up filter/sort render polls

[Aias00]

Related issue

Closes #2023

Summary

useFilterAndSort polled VTable with setInterval(..., 100) until the
rendered CHAT2DB_ROW_NUMBER list changed, then clearInterval. If the
list never changed (empty result set, a filter yielding the same rows, or
the component unmounting mid-poll), the interval ran forever. The
timerId was a local variable inside each handler closure — there was no
ref and no unmount cleanup — so after unmount the interval kept firing and
called setState/table APIs on an unmounted component, leaking memory.
The same pattern appeared at four sites (lines 187/219/250/309).

Changes:

• Add a pollTimerIdsRef that tracks active poll timers.
• Add a pollUntilRowNumbersChange(before, onSettled) helper with a
  MAX_POLL_ITERATIONS cap (100 iterations x 100ms = 10s) so a poll
  cannot run forever when the row list never changes; each timer is
  registered and unregistered on settle/cap.
• Add an unmount useEffect that clears every active timer.
• Route all four inline poll sites (sort, filter clear-all, filter apply,
  clearAllFilters) through the helper.

The cap value (MAX_POLL_ITERATIONS = 100) is chosen generously: the
polls run every 100ms and VTable render settles in milliseconds, so 10s
is far above the normal settle time while still bounding the leak. It is
a single named constant and trivial to tune.

Affected surfaces

• Frontend / Web
• Backend / API / Storage
• Database plugin / Driver
• JCEF / Desktop packaging
• CI / Build / Release
• Documentation only

Verification

• Commands and results:
  • npx tsc --noEmit -p chat2db-community-client/tsconfig.json — no errors reported for useFilterAndSort.tsx.
  • npx eslint src/blocks/SearchResult/components/ResultSetTable/hooks/useFilterAndSort.tsx — no errors.
  • Confirmed only one setInterval remains (the helper's) and all four call sites now call pollUntilRowNumbersChange (lines 224/250/275/328).
• Manual verification: On an empty result set or a same-row filter, a poll now stops after the cap instead of running forever; navigating away mid-poll clears active timers via the unmount effect.
• UI evidence: N/A

Risk and compatibility

• Public API or stored data: N/A.
• Database or driver compatibility: N/A.
• Network, privacy, or security: N/A.
• Community / Local / Pro boundary: N/A.
• Backward compatibility: Polling behavior is unchanged in the normal case (settles, calls sortAfter/filterAfter, clears the timer). Only the previously-leaking paths are bounded and cleaned up.

Reviewer map

• Start here: useFilterAndSort.tsx — new MAX_POLL_ITERATIONS constant (line 33), pollTimerIdsRef + pollUntilRowNumbersChange helper + unmount useEffect (lines 52-83), and the four call sites (224/250/275/328) replacing the inline setInterval blocks.
• Failure condition: Intervals still leak on empty/same-row results or after unmount.
• Rollback or disable path: Revert this single commit; behavior returns to the original (leaking) polls.

Contributor declaration

• I linked the Issue that defines this change.
• I tested the affected behavior and reported the actual results above.
• I did not include credentials, private data, or generated build output.
• I disclosed substantial AI assistance below, or this PR contains no substantial AI-generated code.

AI assistance: The fix, verification, and PR description were produced with Claude Code assistance.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[Aias00]

@openai0229 hi, are you real people? can you add my wechat: aias00

[openai0229]

Reviewed all four poll sites and the timer lifecycle. The bounded polling and Set-based cleanup fix the leak; I also added cleanup when tableInstance is replaced (commit 039924f) so old-instance timers cannot survive without an unmount. Focused ESLint, the combined Community build, and updated required checks pass. No blocking issues remain.