release-21.2: rangefeed: fix panic due to rangefeed stopper race

[erikgrinaker]

Backport 1/1 commits from #76825.

/cc @cockroachdb/release

Release justification: possible node crash.

---

This patch fixes a race condition that could cause an
unexpected Stopped processor panic if a rangefeed registration was
attempted while a store was stopping.

Registering a rangefeed panics if a newly created rangefeed processor is
unexpectedly stopped and the store's stopper is not quiescing. However,
the stopper has two distinct states that it transitions through:
stopping and quiescing. It's possible for the processor to fail to start
because the stopper is stopping, but before the stopper has transitioned
to quiescing, which would trigger this panic.

This patch propagates the processor startup error to the rangefeed
registration and through to the caller, returning before attempting
the registration at all and avoiding the panic. This was confirmed with
50000 stress runs of TestPGTest/pgjdbc, all of which succeeded.

Resolves #76811.
Resolves #76767.
Resolves #76724.
Resolves #76655.
Resolves #76649.
Resolves #75129.
Resolves #64262.

Release note (bug fix): Fixed a race condition that in rare
circumstances could cause a node to panic with
unexpected Stopped processor during shutdown.

---

For details, see #76649 (comment).

[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