fix panic on concurrent map read

[fschoell]

No description provided.

[Copilot]

Pull request overview

This PR addresses a concurrency bug in UpdateCache where concurrent access to the shared entries map across distinct keys could trigger a concurrent map read and map write panic. It does so by introducing an RWMutex to protect map reads/writes while preserving per-key serialization for the update path.

Changes:

• Replaced the cache-wide *sync.Mutex with a cache-wide sync.RWMutex and used RLock/RUnlock for reads in Get.
• Updated Get to safely read the map under a read lock and write under a write lock, preventing concurrent map access panics.
• Added a parallel test that exercises concurrent Get calls across many distinct keys to reproduce/guard against the panic (especially under -race).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
cache.go	Adds an RWMutex and uses it to protect concurrent reads/writes of the entries map in Get (and write-locking in Prune).
cache_test.go	Adds a parallel distinct-keys test to validate the cache no longer panics under concurrent access patterns.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.