Fix AnyKeyspace tablet type selection

[Devanshusharma2005]

Description

This PR fixes a VTGate routing bug where "replica" queries without an explicit keyspace could fail with "no healthy tablet available".
The root cause was that AnyKeyspace() picked the first alphabetical serving keyspace without checking whether it actually had tablets for the requested type. So if the first keyspace only had PRIMARY tablets, an "replica" query would immediately fail even if another keyspace had healthy replicas.
The fix makes global routing tablet-type aware by reusing ResolveDestinations() through a small helper (canResolveKeyspace()). We filter serving keyspaces based on whether they can actually resolve for vc.tabletType, and then continue with the existing selection logic.
If filtering returns nothing, we gracefully fall back to the original behavior to preserve backwards compatibility.
Scope is intentionally small (2 files), no interface changes, and tests cover the main edge cases
Minimal change, aligns global routing with explicit routing behavior.

Related Issue(s)

Fixes : #19243

TEST RESULT

Checklist

• "Backport to:" labels have been added if this change should be back-ported to release branches
• If this change is to be back-ported to previous releases, a justification is included in the PR description
• Tests were added or are not required
• Did the new or modified tests pass consistently locally and on CI?
• Documentation was added or is not required

Deployment Notes

No deployment changes required.
This is purely a VTGate routing logic fix. No changes to topology, VSchema, or tabletmanager.

AI Disclosure

I leveraged perplexity in generating tests. I Manually wrote code using grep commands to find the exact bug file and make changes. Fully understood all changes. @arthurschreiber gave enough hints in the issue.
(gofumpt command is the work of god btw ^^)

[vitess-bot]

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

• Ensure that the Pull Request has a descriptive title.
• Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

• Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

• Apply the release notes (needs details) label if users need to know about this change.
• New features should be documented.
• There should be some code comments as to why things are implemented the way they are.
• There should be a comment at the top of each new or modified test to explain what the test does.

New flags

• Is this flag really necessary?
• Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

• Each item in Jobs should be named in order to mark it as required.
• If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

• Protobuf changes should be wire-compatible.
• Changes to _vt tables and RPCs need to be backward compatible.
• RPC changes should be compatible with vitess-operator
• If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
• vtctl command output order should be stable and awk-able.

[Devanshusharma2005]

@mhamza15 lets have a look.

[mhamza15]

If it's used only within canResolveKeyspace, then it won't cause an allocation. We should colocate it there if it's only used there.

[mhamza15]

We shouldn't use context.Background(). We'll need to funnel in the appropriate context, or at the very least use a short timeout.

[Devanshusharma2005]

Can we do something like this here ? func (vc VCursorImpl) canResolveKeyspace(ksName string) bool {
if vc.resolver == nil {
return true
}
anyShardDestination := []key.ShardDestination{key.DestinationAnyShard{}}
ctx, cancel := context.WithTimeout(context.TODO(), 50time.Millisecond)
defer cancel()
_, _, err := vc.resolver.ResolveDestinations(ctx, ksName, vc.tabletType, nil, anyShardDestination)
return err == nil
} , ig we can use context.TODO when we're unclear and the surrounding is not ready for a ctx param ?

[mhamza15]

Suggested change

	// canResolveKeyspace checks whether the given keyspace has a SrvKeyspace partition
	// for vc.tabletType. Uses ResolveDestinations which reads cached SrvKeyspace data,
	// following the same code path as explicit keyspace routing (Resolver.GetKeyspaceShards).
	// canResolveKeyspace checks whether the given keyspace has a SrvKeyspace partition
	// for vc.tabletType.