Use Fiat-Shamir in IPA

[codygunton]

From: #31 (review): the Fiat-Shamir heuristic must be used to generate the round challenges as part of the ipa_prove and ipa_verify algorithms. The InnerProductArgument class should have a Hash template parameter that defines which hash algorithm is used.

[codygunton]

IPA::reduce_prove should have the same interface as KZG::reduce_prove. In particular, it should consume a ProverOutput of the sort created by Shplonk::reduce_prove

[codygunton]

Perhaps this was handled in a previous PR? Can't quite recall.

[adr1anh]

IPA uses the hash function as the random oracle, which is abstracted away in the Transcript class.
However, it would be a good idea to parametrize the Transcript for different hash functions later on (Poseidon, Reinforced Concrete, Keccack?)