expression: handle duration type infer in least and greatest

[iosmanthus]

What problem does this PR solve?

Issue Number: close #22200

Problem Summary:

What is changed and how it works?

This pull request compares duration as string while they are formalized which will prevent panicking while encountering a non-handled duration type infer.

Related changes

• PR to update pingcap/docs/pingcap/docs-cn:
• Need to cherry-pick to the release branch

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)

Side effects

• Performance regression
  • Consumes more CPU
  • Consumes more MEM
• Breaking backward compatibility

Release note

• expression: handle duration type infer in least and greatest

[iosmanthus]

this pr:

mysql> SELECT LEAST(TIME'00:00:00.1', TIME'00:00:00.12000');
+-----------------------------------------------+
| LEAST(TIME'00:00:00.1', TIME'00:00:00.12000') |
+-----------------------------------------------+
| 00:00:00.1                                    |
+-----------------------------------------------+
1 row in set (0.00 sec)

MySQL:

mysql> SELECT LEAST(TIME'00:00:00.1', TIME'00:00:00.12000');
+-----------------------------------------------+
| LEAST(TIME'00:00:00.1', TIME'00:00:00.12000') |
+-----------------------------------------------+
| 00:00:00.10000                                |
+-----------------------------------------------+
1 row in set (0.01 sec)

Though is' ok and the same as release 4.0, it's better to make it be compatible with MySQL. If it is hard, I think that's fine.

For now, I think we should consider the bug itself, instead of the compatibility with MySQL. We may fix this behaviour in another pull request.

[iosmanthus]

@wjhuang2016 @XuHuaiyu PTAL

[wjhuang2016]

LGTM

[wshwsh12]

Rest LGTM

[wshwsh12]

LGTM

[wshwsh12]

/merge

[ti-srebot]

/run-all-tests

[ti-srebot]

cherry pick to release-4.0 in PR #22421