supernode: consider adding full safe block ref caching to avoid network calls to the EL

1. We should add a failing test which captures the incorrect sync status I would expect the virtual node to return because of this.
2. We then fix the test by having the super authority just return the block ref, and the virtual node calls to the EL to fill in the rest of the information.
3. File an issue to circle back and optimize this, removing the network call by putting some extra caching into the verifier DB.

_Originally posted by @geoknee in https://github.com/ethereum-optimism/optimism/pull/19110#discussion_r2787535400_

No file or test path is named. Start by locating the supernode sync-status test and tracing the super authority and virtual node handling of block references and execution-layer data. Done means the failing test captures the incorrect status, the test passes with the described behavior, and the caching optimization is tracked separately.

Written by the indexing model from the issue text.