release-22.1: opt: fixup CTE stats on placeholder queries

[cucaroach]

Backport 1/1 commits from #100327.

/cc @cockroachdb/release

---

Backport 1/1 commits from #99433.

/cc @cockroachdb/release

---

During optbuilder phase we copy the initial expressions stats into the
fake-rel but this value can change when placeholders are assigned so add
code in AssignPlaceholders to rebuild the cte if the stats change.

Fixes: #99389
Epic: none
Release note: none

[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 

[rytaft]

Reviewed 4 of 4 files at r1, all commit messages.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @cucaroach)

[rytaft]

The only think I'm kind of concerned about this this question from blathers: "What did we do to ensure that a user that doesn’t know & care about this backport, has no idea that it happened?" Stats changes can cause unexpected plan changes even if they are a clear improvement in the general case. Do you think we need a session setting? I'm willing to be convinced we don't, but just want to make sure we thought about it....

[cucaroach]

The only think I'm kind of concerned about this this question from blathers: "What did we do to ensure that a user that doesn’t know & care about this backport, has no idea that it happened?" Stats changes can cause unexpected plan changes even if they are a clear improvement in the general case. Do you think we need a session setting? I'm willing to be convinced we don't, but just want to make sure we thought about it....

Good point! I offer only these weak arguments:

• This went into 22.2 already like this
• It only affects users using recursive CTE prepared statements which seems pretty rare
• I don't feel strongly that this should go into 22.1 and if we're at concerned about it why not limit to 22.2?

[rytaft]

why not limit to 22.2?

Yea, if there's no customer depending on this in 22.1 maybe we can just skip the backport altogether for now? We can always reopen it if needed. Thanks!

[cucaroach]

Closing, per discussion, if customer needs this they can change app to not use placeholders or upgrade to 22.2.