core, eth: implement sparse blobpool

[healthykim]

This is a draft PR for sparse blobpool.

It applies the prototype changes that existed here and here on top of master (with assistance from AI).

[healthykim]

@fjl I am leaving this comment to keep a record of what we discussed today.

Problem: Since we include transactions from the same account in nonce order, if there is a partial transaction, subsequent transactions with higher nonces from the same from account cannot be included by us. I refer to these as blocked transactions (feel free to recommend a better name)

Because a partial transaction can be injected by running multiple nodes above the threshold around the victim, we need a restriction for such transactions. I have two ideas for this.

1. Cap the total size of blocked transactions

Under this policy, we drop transactions whenever size(blockedTxs) > blockedCap || size(txs) > cap (Here blockedCap can be like cap * 0.5). During the eviction, blocked transactions are always considered as cheaper than normal transactions, regardless of the fee they pay. This policy may drop some real valuable transactions, so I am not fully convinced about this part, but on the other hand, I still think it makes sense because blocked transactions are not immediately includable anyway.

2. Fetch the blocking transaction to unblock blocked transactions

Under this policy, when size(blockedTxs) > blockedCap && size(txs) <= cap, we try to change the status of blocked transactions by fetching the full blob of the partial transaction that blocks the subsequent transactions.
We can record something similar to knownTxs to check which peer is serving as a provider for each transaction. If fetching the blocking transaction fails, we drop that transaction and the subsequent blocked transactions. Otherwise it would be okay because now the blocked suffix is shorter than before we fetch the full blob.

I think we can apply both approaches together. I have not implemented the second approach yet. I am working on it butI couldn't find better architecture than adding more dependencies for now.