fix(archiver): skip descendants of invalid-attestations checkpoints

[spalladino]

Motivation

archiver/src/modules/l1_synchronizer.ts skipped checkpoints with insufficient/invalid attestations under the assumption that the next proposer would invalidate them before publishing. When that assumption was violated — i.e., proposer P2 published a valid-attestations checkpoint that extended P1's invalid one — the archiver hit InitialCheckpointNumberNotSequentialError in block_store.addCheckpoints, the catch handler rolled back the L1 sync point, and the next poll re-fetched the same range and re-threw. The archiver looped indefinitely. The protocol already defines OffenseType.PROPOSED_DESCENDANT_OF_CHECKPOINT_WITH_INVALID_ATTESTATIONS for exactly this case but the slasher couldn't see valid-attestations descendants because the archiver threw before emitting any event.

Human Note

This is particularly relevant under pipelining. Attestors now attest to a checkpoint before the previous one is pushed to L1, so they can be inadvertently attesting to a checkpoint built on top of one that became invalid as it was published to the rollup the contract with wrong attestations. So an honest attestor could get slashed if the proposer was malicious.

Approach

In the synchronizer, persist rejected ancestors in the block store keyed by archive root. On each new checkpoint, before attestation validation, compare its header.lastArchiveRoot against the persisted set — if it matches, skip the checkpoint as a descendant of an invalid ancestor and emit a new L2BlockSourceEvents.CheckpointBuiltOnInvalidAncestorDetected event with enough metadata to resolve the proposer. Prune the set by L1 finality so reorg-vulnerable rows are dropped only once they can no longer be displaced. The slasher's AttestationsBlockWatcher is fixed to slash the proposer (not the attestors) under the existing offense type and gains a new handler for the new event.

Changes

• stdlib: New L2BlockSourceEvents.CheckpointBuiltOnInvalidAncestorDetected enum value and CheckpointBuiltOnInvalidAncestorEvent payload (committee/seed/epoch/checkpoint/ancestor info).
• archiver: New RejectedCheckpoint type and LMDB-backed store API in block_store.ts (getRejectedCheckpoints, addRejectedCheckpoint, removeRejectedCheckpointsBelowL1Block). l1_synchronizer.ts learns Branch A (descendant skip — emit event, persist, continue) ahead of attestation validation, and prunes rejected rows below the finalized L1 block at the end of each sync iteration. Added getEpochInfoForCheckpoint helper in validation.ts so the descendant path can populate the event payload without paying for signature verification twice.
• slasher: Existing slashAttestorsOnAncestorInvalid renamed to slashProposerOnAncestorInvalid and switched from attestor list to a single proposer (resolved the same way slashProposer does), fixing the inconsistency between the offense's proposer-oriented naming and the previous attestor-slashing behavior. New handleDescendantOfInvalid listens to the new event and slashes the proposer of the valid descendant. Class JSDoc updated to describe both paths.
• archiver (tests): New block_store.test.ts cases for the rejected-checkpoints store. archiver-sync.test.ts gains a regression for the valid-descendant skip and updates the existing invalid-attestations test now that descendants also persist.
• slasher (tests): Existing descendant test updated to expect proposer-slash (one emit, not two attestor emits); new cases cover the new event handler, cascading descendants, deduplication, and the no-proposer fallback.
• end-to-end (tests): The e2e regression introduced in this branch is flipped from asserting the stuck-state bug to asserting the post-fix behavior — the chain advances past the bad descendant, the new event fires, and both offenses (PROPOSED_INSUFFICIENT_ATTESTATIONS for P1's proposer and PROPOSED_DESCENDANT_OF_CHECKPOINT_WITH_INVALID_ATTESTATIONS for P2's proposer) are recorded.
• sequencer-client: Test-only skipWaitForValidParentCheckpointOnL1 flag (from the original test commit) bypasses the parent-validity check inside CheckpointProposalJob.waitForValidParentCheckpointOnL1 so the e2e can engineer the fault.