Cache subnetIDs in cachedState

[michaelkaplan13]

Why this should be merged

Optimizes look ups of the subnet ID for a given blockchain ID, particularly for chains with validatorState instances that would otherwise require going over grpc for the look up each time (i.e. subnet-evm instances).

See discussion here.

How this works

Uses an LRU cache for subnet IDs for given blockchain IDs.

How this was tested

Unit tests

Need to be documented in RELEASES.md?

No

[Copilot]

Pull Request Overview

Adds an LRU cache for subnet IDs keyed by blockchain ID to avoid repeated underlying (potentially gRPC) lookups and renames the existing validator set cache field for clarity. Also introduces unit tests validating cache population, hits, and error handling paths.

• Introduces subnetIDsCache (LRU) with configurable size.
• Renames internal validator set cache field and adds GetSubnetID caching logic.
• Adds comprehensive test for subnet ID caching behavior including error path.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
snow/validators/state.go	Adds subnet ID cache, renames validator set cache field, implements cached GetSubnetID.
snow/validators/state_test.go	Adds unit test covering subnet ID cache miss/hit and error handling behavior.

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[michaelkaplan13]

@StephenButtolph called out that this does not actually improve the worst case verification times/cost for flows dependent on GetSubnetID, since bogus blockchain ID parameters could be provided, which will never be cached.

It's strictly a best case performance optimization, but should not be relied upon for gas price calculations.