Unnecessary block serialized byte comparison

Everything after the hash comparison is unnecessary because the hash is constructed by rlp encoding the block.

```
func ensureBlocksMatch(espresso, fullNode *types.Block) error {
if espresso.Hash() != fullNode.Hash() {
return fmt.Errorf("block hash mismatch: espresso %s, full node %s", espresso.Hash(), fullNode.Hash())
}

espressoRLP, err := rlp.EncodeToBytes(espresso)
if err != nil {
return fmt.Errorf("failed to RLP-encode espresso block: %w", err)
}
fullNodeRLP, err := rlp.EncodeToBytes(fullNode)
if err != nil {
return fmt.Errorf("failed to RLP-encode full node block: %w", err)
}
if !bytes.Equal(espressoRLP, fullNodeRLP) {
return fmt.Errorf("block mismatch at number %d: espresso and full node blocks differ", espresso.NumberU64())
}
return nil
}
```

Search for the ensureBlocksMatch function and read its callers to confirm the hash comparison is the relevant validation entry point. Remove the redundant serialized-byte comparison while preserving the existing hash-mismatch error behavior, then verify the project tests pass.

Written by the indexing model from the issue text.