apollo_batcher: read state commitment infos from the committer on a cache miss

[yoavGrs]

No description provided.

[reviewable-StarkWare]

This change is 

[yoavGrs]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• apollo_storage: drop the legacy state commitment infos table and file at startup #15075
• apollo_storage,apollo_batcher,apollo_reverts: remove state commitment infos from storage #15068
• apollo_batcher: read state commitment infos from the committer on a cache miss #15067 👈 (View in Graphite)
• apollo_batcher: evict the reverted height from the caches after writing the pending results #15076
• apollo_batcher: cache recent state commitment infos in memory #15066
• apollo_committer,apollo_committer_types: add an API for reading state commitment infos #15065
• main-v0.14.4

This stack of pull requests is managed by Graphite. Learn more about stacking.

[cursor]

PR Summary

Medium Risk
Adds a committer dependency at batcher startup for cache warming; failures are non-fatal but may increase storage reads until commitments catch up.

Overview
On batcher start, the node now prefetches state commitment infos from the committer for recent heights already committed in storage (from storage_height - STORED_BLOCK_HASH_BUFFER up to the commitment task offset) and fills recent_state_commitment_infos_cache, so get_state_commitment_infos / has_state_commitment_infos can hit memory instead of storage for that window.

If the committer RPC fails, startup continues with an empty cache (warning only). Default test mocks stub get_state_commitment_infos; new tests assert the requested height range and tolerance for committer outages.

^{Reviewed by Cursor Bugbot for commit 1d4d2db. Bugbot is set up for automated code reviews on this repo. Configure here.}