release-21.1: optbuilder: do not create invalid casts when building CASE expressions

[mgartner]

Backport 1/1 commits from #76193.

/cc @cockroachdb/release

---

The optbuilder no longer creates invalid casts when building CASE
expressions that have branches of different types. CASE expressions that
previously caused internal errors now result in user-facing errors. A
similar change was made recently to UNION expressions in #75219.

Note that there is more work to be done to be fully consistent with
Postgres's CASE typing behavior, see #75365.

Fixes #75365

Release note (bug fix): CASE expressions with branches that result in
types that cannot be cast to a common type now result in a user-facing
error instead of an internal error.

---

Release justification: This fixes a minor bug that causes an internal error.

[blathers-crl]

Thanks for opening a backport.

Please check the backport criteria before merging:

• Patches should only be created for serious issues or test-only changes.
• Patches should not break backwards-compatibility.
• Patches should change as little code as possible.
• Patches should not change on-disk formats or node communication protocols.
• Patches should not add new functionality.
• Patches must not add, edit, or otherwise modify cluster versions; or add version gates.

If some of the basic criteria cannot be satisfied, ensure that the exceptional criteria are satisfied within.

• There is a high priority need for the functionality that cannot wait until the next release and is difficult to address in another way.
• The new functionality is additive-only and only runs for clusters which have specifically “opted in” to it (e.g. by a cluster setting).
• New code is protected by a conditional check that is trivial to verify and ensures that it only runs for opt-in clusters.
• The PM and TL on the team that owns the changed code have signed off that the change obeys the above rules.

Add a brief release justification to the body of your PR to justify this backport.

Some other things to consider:

• What did we do to ensure that a user that doesn’t know & care about this backport, has no idea that it happened?
• Will this work in a cluster of mixed patch versions? Did we test that?
• If a user upgrades a patch version, uses this feature, and then downgrades, what happens?

[cockroach-teamcity]

This change is 

[RaduBerinde]

Reviewable status: complete! 1 of 0 LGTMs obtained