expression: avoid constPropgate for the hybrid types (#20258)

[ti-srebot]

cherry-pick #20258 to release-3.0

---

What problem does this PR solve?

Issue Number: close #20180

Problem Summary:

create table t(a enum('a', 'b'), b tinyint);
create table t1(c varchar(20));
insert into t values('b', 0);
insert into t1 values('b');
select * from t, t1 where t.a = t1.c and t.b= t1.c;

In condition t.a=t1.c, the EvalType is ETString.
In condition t.b=t1.c, the EvalType type is ETInt.
So that it shouldn't use propagate column optimization, only if the propConstSolver can distinguish them.
In the case above, it will construct a new condition t.a=t.b and push down to table t, and make the join result wrong.

What is changed and how it works?

Proposal: xxx

What's Changed: avoid propagate column EQ for Hybrid type (enum,set,bit).

How it Works:

Related changes

• Need to cherry-pick to the release branch 3.0,4.0

Check List

Tests

• Unit test
• Integration test

Side effects

• Performance regression
  • Consumes more CPU
  • Consumes more MEM

Release note

• Fix potentially incorrect results when one of the equal condition is Enum of Set type.

[ti-srebot]

/run-all-tests

[eurekaka]

Please resolve conflicts.

[lzmhhh123]

LGTM

[ti-srebot]

@lzmhhh123, Thanks for your review, however we are sorry that your vote won't be count.

[eurekaka]

LGTM

[ti-srebot]

@eurekaka, Thanks for your review. The bot only counts LGTMs from Reviewers and higher roles, but you're still welcome to leave your comments. See the corresponding SIG page for more information. Related SIG: execution(slack).

[wshwsh12]

lgtm

[bb7133]

/merge

[ti-srebot]

Sorry @bb7133, this branch cannot be merged without an approval of release maintainers

[bb7133]

/run-all-tests

[XuHuaiyu]

LGTM

[XuHuaiyu]

/merge

[ti-srebot]

/run-all-tests

[ti-srebot]

@ti-srebot merge failed.

[bb7133]

/run-unit-test