refactor: add from bytes trait to content key

[njgheorghita]

What was wrong?

Minor refactor to content keys

How was it fixed?

Added From trait to simplify construction

To-Do

• Clean up commit history and use conventional commits.

[KolbyML]

This refactor or at least the idea of it from my perception came from reviewing my PR #1436 . Well I don't think this PR is that related directly to #1436 other then touching a similar area of code, I think #1436 should just be reviewed, I think offshoot PR's shouldn't be made until the original PR being reviewed is reviewed. This kind of seems like leap frogging in a sense, when I think the priority should be getting #1436 reviewed and merged in the first place.

[njgheorghita]

Yup, i'm in the process of reviewing #1436, it's just a pretty large pr, don't mind waiting to merge this until that's through

[morph-dev]

While I'm not against this, I personally like more explicit approach, something like this:

impl HistoryContentKey {
    pub fn new_block_header_with_proof(block_hash: impl Into<[u8; 32]>) -> Self {
        Self::BlockHeaderWithProof(BlockHeaderKey {
            block_hash: block_hash.into(),
        })
    }

    pub fn new_block_body(block_hash: impl Into<[u8; 32]>) -> Self {
        Self::BlockBody(BlockBodyKey {
            block_hash: block_hash.into(),
        })
    }

    pub fn new_block_receipts(block_hash: impl Into<[u8; 32]>) -> Self {
        Self::BlockReceipts(BlockReceiptsKey {
            block_hash: block_hash.into(),
        })
    }

    ... // already existing functions
}

I'm fine with your approach as well (or both), so pick whatever you prefer.

🚢

[KolbyML]