Could the trie root of the mine workflow be optimized using a sparse tree?

[will-2012]

reth/crates/evm/evm/src/execute.rs

Lines 516 to 520 in 8f5b4ae

	// calculate the state root
	let hashed_state = state.hashed_post_state(&db.bundle_state);
	let (state_root, trie_updates) = state
	.state_root_with_updates(hashed_state.clone())
	.map_err(BlockExecutionError::other)?;

The code location is as above.

[will-2012]

Could the current trie-root be made more pluggable to facilitate its use with miner.

It's noticeable that the current sparse-tree seems to be used only for synchronization. Do the team have any suggestions for using the sparse-tree when the block builder produces blocks? (Currently, the sparse-tree relies heavily on the internal state of the engine-tree and is deeply bound to the payload-proprocesor; it doesn't seem very reusable.)

[will-2012]

Any help or suggestions regarding this issue would be greatly appreciated! :D.