release-22.1: kvserver: disable merge queue until kvsubscriber has updated

[blathers-crl]

Backport 1/1 commits from #78122 on behalf of @irfansharif.

/cc @cockroachdb/release

---

If we don't have any span configs available, enabling range merges would
be extremely dangerous -- we could collapse everything into a single
range. We've observed this happen when the kvsubscriber's initial scan
overflows its bounded buffer, preventing it from ever getting a
snapshot. A future commit will fix the bounded memory issue, but the
side-effect pointed out the need for this important safe guard.

Informs #77687.

Release justification: bug fix
Release note: None

---

Release justification:

[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 

[nvanbenschoten]

LGTM. Do you think we need similar protection in any of the other queues?

[irfansharif]

The split queue will implicitly not be able to do anything, if the kvsubscriber doesn’t have data, the split queue doesn’t have any split points it can induce. Only other queue I can think of is maybe the MVCC GC queue and it not having protected timestamps installed by SQL. I’m not sure if holding up GC is the right solution there. @arulajmani, what do you think?

[arulajmani]

I think it's okay to not hold up GC if protected timestamps haven't been reconciled. All our claims are for when protected timestamps are reconciled to KV, not when they're written in SQL, so we should be good.

[ajwerner]

But this happens after reconciliation

[arulajmani]

Ah right -- I'll send out a patch that doesn't allow GC-ing if PTS information isn't available.

[arulajmani]

#78300