Support utilizing merkle tree for AddressAllowlistCondition

[Muhammad-Altabba]

I feel like a merkle tree would be useful here. Condition/Message Kit would only need to store the merkle root.

I think this is a great suggestion. But it would need us to discuss more on where to save the full Merkle tree from which the decryptor can take just his relevant proof branches and send it to the nodes. As the full tree would not be sent on the nodes nor it would be part of the cyphertext, I guess. Also its implementation, would I guess, exceed the initial scope of this feature and make it take longer...

If we would have a usecase that needs to support a long list of the allowlist, a Merkle tree is the one to go. But we need then to discuss the data structure. As I think that the decryptor would need to know the Merkle tree path related to his address and pass it to the nodes as part of his proof.

Originally posted by @Muhammad-Altabba in https://github.com/nucypher/taco-web/pull/665#discussion_r2153188508