Caplin: support backfilling blobs through a remote beacon API.

[Giulio2002]

Adds support for backfilling historical blob data through a remote Beacon API as an alternative to P2P synchronization.

Implementation

• BeaconApiBlobDownloader: Fetches data column sidecars via Beacon API (/eth/v1/beacon/data_column_sidecars/{slot}) using SSZ encoding, stores columns, and triggers blob recovery via PeerDAS when sufficient columns are available
• BlobHistoryDownloader: Refactored P2P-based blob backfiller into standalone component implementing common BlobBackfiller interface
• Automatic selection: Stages layer selects Beacon API downloader when --caplin.blobs-backfiller-url is set, otherwise falls back to P2P

Configuration

# Use remote Beacon API for blob backfilling
erigon --caplin.blobs-backfiller-url=http://beacon-node:5052 --caplin.blobs-archive

# Alternative: immediate backfill of recent 4096 epochs
erigon --caplin.blobs-immediate-backfill

# Disable blob pruning for archival
erigon --caplin.blobs-no-pruning

Data Column Integration

• PeerDAS manages data column storage and blob recovery
• When ≥50% of columns are available for a block, schedules blob reconstruction
• Supports both direct blob download and column-based recovery paths

Backfill Strategy

Both downloaders work backwards from current slot to Deneb fork epoch, skipping slots where:

• Blobs already exist in storage
• Block is pre-Deneb or has no blob commitments
• Slot is covered by frozen snapshots

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[Giulio2002]

@copilot please make a description of the PR and post it as a comment, I am too lazy to write it myself

[Copilot]

@Giulio2002 I've opened a new pull request, #18344, to work on those changes. Once the pull request is ready, I'll request review from you.