remove lru cache for store_adapter

Author: julienrbrt

CHANGELOG.md

@@ -17,6 +17,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Improve `cache.NumPendingData` to not return empty data. Automatically bumps `LastSubmittedHeight` to reflect that. ([#3046](https://github.com/evstack/ev-node/pull/3046))
 - **BREAKING** Make pending events cache and tx cache fully ephemeral. Those will be re-fetched on restart. DA Inclusion cache persists until cleared up after DA inclusion has been processed. Persist accross restart using store metadata. ([#3047](https://github.com/evstack/ev-node/pull/3047))
+- Replace LRU cache by standard mem cache with manual eviction in `store_adapter`. When P2P blocks were fetched too fast, they would be evicted before being executed [#3051](https://github.com/evstack/ev-node/pull/3051)
 
 ## v1.0.0-rc.2
 
@@ -25,7 +26,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Improve cache handling when there is a significant backlog of pending headers and data. ([#3030](https://github.com/evstack/ev-node/pull/3030))
 - Decrease MaxBytesSize to `5MB` to increase compatibility with public nodes. ([#3030](https://github.com/evstack/ev-node/pull/3030))
 - Proper counting of `DASubmitterPendingBlobs` metrics. [#3038](https://github.com/evstack/ev-node/pull/3038)
-- Replace `go-header` store by `ev-node` store. This avoid duplication of all blocks in `go-header` and `ev-node` store. Thanks to the cached store from #3030, this should improve p2p performance as well.
+- Replace `go-header` store by `ev-node` store. This avoid duplication of all blocks in `go-header` and `ev-node` store. Thanks to the cached store from #3030, this should improve p2p performance as well. [#3036](https://github.com/evstack/ev-node/pull/3036)
 
 ## v1.0.0-rc.1