planner: join resolveIndex won't find its column and amend join's l/rused logic for reversed column ref from join schema to its children

[ti-chi-bot]

This is an automated cherry-pick of #51203

What problem does this PR solve?

Issue Number: close #42588

Problem Summary:

What changed and how does it work?

issue #42588 is caused by many factors and this pr is try to resolve them all.

• first: resolve parent pointer modification error caused by the apply elimination, as mentioned here Panic triggered at chunk.appendCellByCell (util/chunk/chunk.go:421)  #42588 (comment)
• second: since TiDB internal has many non-rigorous logical optimizing rules, occasionally two or more of them will leading a critical assumption which will lead a resolveIndex error as mentioned here: planner: refactor Join and Limit's ResolveIndices #45831 (comment)
• third: in the execution of join, the inline projection's lused and rused are order unawareness, which means even if we allow reversed column ref in join schema itself to its children, it will break its assumption here too. So we need to fix here as well.
  example:

join schema[col11, col22, col21], child1 [col1], child2 [col21, col22]
the pull/45831 make an assumption here that join schema won't refer a column-order revered from the order of its children. Like here [col22, col21] in join schema is reversed with  [col21, col22] from its child2. Step2 try to fix this.

in the execution layer.
join schema[col11, col22, col21], child1 [col1], child2 [col21, col22]
the lused will be [true]
the rused will be [true, true]
when constructing the join row result, the order couldn't be restored from a simple bool array. Consequently one join result will be shown like: <col1, col21, col22> while the field type is totally different from what the datum in the same position. So step3 try to fix this.

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

[fixdb]

+1

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: AilinKid, fixdb

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

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [AilinKid]

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:

• 2024-02-26 08:05:51.439038727 +0000 UTC m=+862840.186661838: ☑️ agreed by fixdb.
• 2024-02-27 08:30:05.281655436 +0000 UTC m=+950694.029278549: ☑️ agreed by AilinKid.

[AilinKid]

/retest

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (release-7.1@ff7d92f). Click here to learn what that means.
The diff coverage is 75.1445%.

Additional details and impacted files

@@               Coverage Diff                @@
##             release-7.1     #51282   +/-   ##
================================================
  Coverage               ?   73.4102%           
================================================
  Files                  ?       1210           
  Lines                  ?     379064           
  Branches               ?          0           
================================================
  Hits                   ?     278272           
  Misses                 ?      83004           
  Partials               ?      17788