release-22.1: vendor: bump cloud.google.com/go/storage from v18.2.0 to v1.21.0

[adityamaru]

This commit bumps the cloud.google.com/go/storage vendor to
include the ability to inject custom retry functions when reading
and writing from the underlying SDK -
https://pkg.go.dev/cloud.google.com/go/storage#Client.SetRetry.

The motivation for this change is to combat the high rate of
restores we are seeing fail due to an internal http2 stream error
that is being surfaced by the SDK in our roachtests. As seen in #85024
we would like to wrap the default retry logic with our custom retry
handling for this particular error. This is the recommended solution as per:
googleapis/google-cloud-go#3735
googleapis/google-cloud-go#784

Note, the dependencies have been bumped to the version that we have
been running on master since the 22.1 branch was cut.

Release note (general change): bump cloud.google.com/go/storage from
v18.2.0 to v1.21.0 to allow for injection of custom retry logic in the
SDK

Release justification: necessary dependency bump that allows us to backport a fix
to prevent restores from failing with an internal stream HTTP2 error as seen in
several roachtests. The fix that we wish to backport is #85024.

[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 

[adityamaru]

We have seen a support escalation with this error too - https://cockroachdb.zendesk.com/agent/tickets/13190 and discussed internally that we would like to consider a backport for #85024 given that its early days for 22.1. This change does bump:

google.golang.org/grpc v1.46.0
google.golang.org/protobuf v1.28.0

as well.

[knz]

I've reviewed manually the protobuf and grpc changes and those two LGTM.

[stevendanna]

For protobufs, the only change that stood out to me was the recursion limit change: https://github.com/protocolbuffers/protobuf-go/releases/tag/v1.28.0 but the default limit of 10000 seems high enough to avoid any practical problems.

For gRPC, the only change that jumped out to me was the status code change here https://github.com/grpc/grpc-go/releases/tag/v1.45.0 but I believe we fixed the only known issue with that here: aea1915

[adityamaru]

Thanks for the reviews!