Consider moving `verify_multi_opening` outside of FK20

[kevaundray]

This has been taken from Benedikt Wagner's document.

There is a comment here re why its currently there:

rust-eth-kzg/cryptography/kzg_multi_open/src/fk20/verifier.rs

Line 110 in e043452

/// Note: Although this method is on the `FK20Verifier` structure, it is possible to verify methods that are not

Related to #246

I'm still on the fence about it from a code organization perspective -- note that all of our "internal naive" provers have their own verifiers for simplicity. Though the complexity might just be an artifact of the current architecture. We could have a "provers" and "verifiers" folder for example, and the verifiers are just naive and optimized.

[b-wagn]

We could have a "provers" and "verifiers" folder for example, and the verifiers are just naive and optimized.

That sounds reasonable to me 👍 But I understand that it requires a non-trivial restructuring of the code.