Fix getAccountProof for Bonsai

[gfukushima]

PR description

This PR fix the eth_getProof RPC method when using Bonsai by adding the worldStateProof to BonsaiWorldStateProvider.
By passing a snapshot of the ws that we want the proof for the BonsaiWorldStateProvider should be able to make normal usage of the worldStateProof to generate the proofs.

Fixed Issue(s)

Part of #5377

[github-actions]

• I thought about documentation and added the doc-change-required label to this PR if updates are required.
• I thought about the changelog and included a changelog update if required.
• If my PR includes database changes (e.g. KeyValueSegmentIdentifier) I have thought about compatibility and performed forwards and backwards compatibility tests

[matkt]

seems to be good

do we can add test for bonsai in this class WorldStateProofProviderTest. the current tests are only using forest

you can use this class as example SnapWorldDownloadStateTest we are using Parameterized to call the same tests with bonsai and forest

[siladu]

Suggested change

	if (dataStorageFormat.equals(DataStorageFormat.BONSAI)) {
	worldStateStorage =
	new BonsaiWorldStateKeyValueStorage(
	new InMemoryKeyValueStorageProvider(), new NoOpMetricsSystem());
	} else {
	worldStateStorage = new WorldStateKeyValueStorage(new InMemoryKeyValueStorage());
	}
	worldStateStorage =
	switch (dataStorageFormat) {
	case BONSAI -> new BonsaiWorldStateKeyValueStorage(
	new InMemoryKeyValueStorageProvider(), new NoOpMetricsSystem());
	case FOREST -> new WorldStateKeyValueStorage(new InMemoryKeyValueStorage());
	};

[siladu]

Why is this necessary?

[matkt]

we have to save the state root and the blockhash in specific keys. when we are doing an isWorldStateAvailable it is this key that are checked. So if this keys are empty the isWorldStateAvailable will return false

[siladu]

Maybe worth a comment explaining this?
Presume it's a bonsai-only issue?

[matkt]

can we verify this GetProof method (compare with other client ?) from what i see getProof doesn't look good. I worked on this part a long time ago but I have a doubt now

Normally GetProof must return the siblings at each level in order to validate the proof. That doesn't seem like it regarding the code. I implemented a getProof method in another component so we can fix that easily but just wanted to be sure i'm not missing something

IMO it should be L1+L2+X for X

*         o
*        / \
*       o   o (L1)
*       / \
*  (L2)o  x

looking at the code it is L1+L2+X

*   (L1)  o
*        / \
*   (L2)o   o 
*       / \
*      o   x 

[gfukushima]

I've tested and we do provide the right proofs but there's seems to be an issue for proofs there are not the latest state in bonsai. Will move this PR to draft until I get back to it.

[gfukushima]

Closing this until I get back to it.