release-21.2: sql: use clear range to clean tables during rollback of create

[fqazi]

Backport 1/1 commits from #79543.

/cc @cockroachdb/release

---

Fixes: #79123

Previously, when dropping a table during rollback of a create
we would clean up table data in chunks instead of doing a
clear range. This was inadequate because he code path in
question existed for compatibility Cockroach 1.1 and had terrible
performance characteristics for large tables. To address,
this patch will always use delete data using clear range when
dropping tables during rollback which will have better performance.

Release note (performance improvement): Rollback of CREATE
TABLE AS with large quantities of data has similar performance
to regular DROP TABLE.

Release justification: low risk since it's used during rollback when a table will be deleted and exploits existing logic we should have been using.

[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 

[fqazi]

@postamar TFTR!