Nondegenerate multisignature hardware wallets

[murisi]

Describe your changes

• Combined Signature and MultiSignature to reduce logic duplication. Specifically:
  • Decompression logic no longer needs to be separately implemented for both types of signatures
  • Hardware wallet would no longer need to implement separate hashers, storage logic, and serializers for both types of signatures
• Allow signer to be specified in signature section, otherwise indexing data would need to be sent to hardware wallet for nondegenerate multisig transactions
• Added decompression logic in order to eliminate the decompression hacks we are using in namada-interface (we should explore in future whether this can be used to reduce size of signature sections on protocol)
• Using map instead of set to store indexed signatures in order to remedy the signature double counting issues (multiple valid signatures can exist for given hash and public key).
• Eliminated signature double counting across different sections by storing successfully verified indices in a set.

Indicate on which release or other PRs this topic is based on

Namada v0.22.0

Checklist before merging to draft

• I have added a changelog
• Git history is in acceptable state

[grarco]

I think that given this rework it would be slightly better to move the gas accounting part directly into the verify_signature method of Signature, to have a highest possible resolution. But we can do this later

[grarco]

Looks good to me, thanks!