[DRAFT] feat: pqsig specs for devnet1

[unnawut]

Opening this PR prematurely for discussion

🗒️ Description

Listing down expected changes if we introduce SignedMessageBundle to allow a single PQ key to authenticate multiple actions in a slot. The general direction from PQinterop keygen session.

We can take another path and just keep the signed messages separate i.e. SignedBlock, SignedVote but need more discussion on requiring multiple PQkeys or whether 1 action per slot is really acceptable.

Container changes

• Modify: State::validators: { pubkeys: xmss::containers::PublicKey }
• Modify: BlockBody::attestations: List[SignedMessageBundle]
• Add SignedMessageBundle : { block_message: Block, vote_data: Vote, signature: Signature }
• Add: Signature: { path: HashTreeOpening, rho: Randomness, hashes: List[HashDigest] } -> Added but not
• Remove: SignedBlock -> use SignedMessageBundle instead
• Remove: SignedVote -> use SignedMessageBundle instead

STF changes

• Modify: state_transition from receiving signed_block to block and block signature pre-validated
• Add: process_bundle(signed_message_bundle)
• ❓ Modify: process_attestations()from receiving attestations: List[SignedVote] to List[SignedMessageBundle] (TBD)

Gossipsub

• ❓ Add: SignedMessageBundle
• ❓ Remove: SignedBlock
• ❓ Remove: SignedVote

Note: this looks worse for network optimization

ResResp

• ❓ Modify: BlocksByRootResponse ReqResp returns List[SignedBlock] -> returns List[SignedMessageBundle]. Ideal?

✅ Checklist

• Ran tox checks to avoid unnecessary CI fails:

  uvx --with=tox-uv tox

• Considered adding appropriate tests for the changes.
• Considered updating the online docs in the ./docs/ directory.

[unnawut]

Closing in favor of #67