stats: add defensive guard for interval calculation

[0xPoe]

What problem does this PR solve?

Issue Number: close #65815

Problem Summary:
Negative durations from mysql.analyze_jobs (e.g. clock skew/bad records) can happen and should be handled defensively without skewing retry logic.

What changed and how does it work?

Based on #65821.

• Document why negative average durations return NoRecord.
• Add unit tests for negative finished-job durations.

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No need to test

  • I checked and no code files have been changed.

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

None

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 78.6447%. Comparing base (258332a) to head (1365b14).
⚠️ Report is 22 commits behind head on master.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #65889        +/-   ##
================================================
+ Coverage   77.7820%   78.6447%   +0.8626%     
================================================
  Files          2001       1932        -69     
  Lines        545618     537652      -7966     
================================================
- Hits         424393     422835      -1558     
+ Misses       119563     114372      -5191     
+ Partials       1662        445      -1217     

Flag	Coverage Δ
integration	44.6222% <0.0000%> (-3.5377%)	⬇️
unit	77.0866% <100.0000%> (+0.6672%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
dumpling	56.7974% <ø> (ø)
parser	∅ <ø> (∅)
br	48.8835% <ø> (-12.0906%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[0xPoe]

🔢 Self-check (PR reviewed by myself and ready for feedback)

• Code compiles successfully

• Unit tests added

• All tests pass

• Bazel files updated

• Comments added where necessary

• PR title and description updated

• Documentation PR created (or confirmed not needed)

• PR size is reasonable

/cc @terry1purcell @henrybw

[Copilot]

Pull request overview

This PR hardens the auto-analyze priority queue’s interval calculations against bad or skewed mysql.analyze_jobs duration records, preventing negative durations from corrupting retry/skip logic. It also clarifies the NoRecord sentinel semantics and adds regression tests around negative durations.

Changes:

• Update GetAverageAnalysisDuration and GetLastFailedAnalysisDuration to treat negative durations as non-usable (either NoRecord or a bounded default wait time) instead of flowing them into scheduling logic.
• Switch last-failed duration reading to a signed integer and document that both functions return NoRecord when no usable history exists.
• Add unit tests that insert synthetic negative-duration records to validate the new defensive behavior for both average and last-failed analysis durations.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
pkg/statistics/handle/autoanalyze/priorityqueue/interval.go	Adds defensive guards for negative average/failed analysis durations, uses signed integer for TIMESTAMPDIFF results, and updates comments to state that NoRecord is returned when there is no usable history.
pkg/statistics/handle/autoanalyze/priorityqueue/interval_test.go	Extends tests to cover negative finished and failed job durations, ensuring they map to NoRecord or the default failed-analysis wait time instead of influencing skip logic.

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: henrybw, terry1purcell

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• pkg/statistics/OWNERS [terry1purcell]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ti-chi-bot]

[LGTM Timeline notifier]

Timeline:

• 2026-02-02 17:14:27.915241804 +0000 UTC m=+114339.016640523: ☑️ agreed by terry1purcell.
• 2026-02-02 18:51:18.657019501 +0000 UTC m=+120149.758418220: ☑️ agreed by henrybw.

[0xPoe]

/retest

[0xPoe]

/test all

[0xPoe]

/test all

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-8.5: #66009.