release-21.1: opt: adjust cost of scan with unbounded cardinality to avoid bad plans

[rytaft]

Backport 1/1 commits from #68676.

/cc @cockroachdb/release

Release justification: This is a minor tweak to the cost model that fixes a recent regression caused by #67388. The regression was affecting the plans of customers, so this backport is needed to fix the plan regressions.

---

This commit tweaks the application of the unbounded cardinality penalty
in the coster to add it directly to the cost of scans and zigzag joins
rather than to the row count. This helps prevent an issue in which the
number of index columns could dominate the scan cost and result in
suboptimal plans.

Fixes #68556

Release note (bug fix): Fixed a regression in the optimizer's cost model
that could cause it to choose suboptimal plans when choosing between two
non-unique index scans with different numbers of columns per index.

[blathers-crl]

Thanks for opening a backport.

Please check the backport criteria before merging:

• Patches should only be created for serious issues.
• 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.

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]

Due to the somewhat urgent nature of this issue, I do not think it should bake for the full two weeks. It's already been on master for almost a week without issues, so I will merge this as soon as it's approved.