Decide whether to keep or remove RPO/RPX hash functions

[huitseeker]

We need to decide if RPO and RPX should be kept or deleted from the codebase.

Discussion: #720 (comment)

Current status: keeping them for now, but should formally decide if they'll be removed in a future release.

Related to #720

[huitseeker]

Additional context from PR review:

The RandomCoin structs (RpoRandomCoin and RpxRandomCoin) are also under consideration:

• Plonky3-rebased (alternate) #720 (comment) (already linked in issue description)
• Plonky3-rebased (alternate) #720 (comment)

The main use case for RandomCoin structs was in Winterfell proving/verifying. With the move to Plonky3, they may no longer be needed.

Decision needed:

• Should RPO/RPX hash functions be kept or removed?
• Should RpoRandomCoin/RpxRandomCoin be kept or removed?
• If keeping for transition period, what's the timeline for removal?

[Al-Kindi-0]

I think it is pretty early/unclear whether we want to keep or remove RPO/RPX hash functions from miden-crypto.
Regarding the RandomCoin I think once we transition to P3 they are obsolete.

[bobbinth]

I think the first step here is to make Poseidon2 our primary hash function. This would involve moving all data structures (e.g., Merkle-related structs) to Poseidon2.

[adr1anh]

We should be able to use DuplexChanllenger though I haven't fully looked into the details.

[Al-Kindi-0]

We should be able to use DuplexChanllenger though I haven't fully looked into the details.

Not sure I follow here, this is what we are currently using for implementing the Fiat-Shamir part of the protocol.