Add test for BabyBearDigest

[hadjiszs]

Would be nice to have a test for this checking the implementation against the reference in sp1. Could make the BabyBearDigest implement sp1_prover::HashableKey in test mode and compare the output of this to the output of the default impl of the trait method. It could even be a randomized test as it's supposed to work for any valid BabyBearDigest.

Originally posted by @iljakuklic in #186 (comment)

[hadjiszs]

However my favorite option here would be to remove the need for this in the prover by having in the L1 directly the hash_u32 stored.

I think that the fact that in the L1 the hash_bn254 is used is only inherited from op which verifies this aggregation proof as plonk proof on the L1, which we don't

cc @Freyskeyd @invocamanman