core: implement in-block prefetcher

[rjl493456442]

This pull request enhances the block prefetcher by executing transactions in parallel
to warm the cache alongside the main block processor.

Unlike the original prefetcher, which only executes the next block and is limited to chain
syncing, the new implementation can be applied to any block. This makes it useful not
only during chain sync but also for regular block insertion after the initial sync.

TODO

• experiment whether the state hashing is necessary in block prefetcher (duplicate with trie prefetcher)

[rjl493456442]

PR: bench05
Master: bench06

• PR is about 10% faster than Master
• The speedup comes from the faster account/storage read
• The memory allocation and CPU usage is about 2x than master

File: geth
Type: inuse_space
Time: 2025-03-31 19:57:06 CST
Entering interactive mode (type "help" for commands, "o" for options)
(pprof) alloc_space
(pprof) top
Showing nodes accounting for 3139.56GB, 49.79% of 6305.50GB total
Dropped 2237 nodes (cum <= 31.53GB)
Showing top 10 nodes out of 245
      flat  flat%   sum%        cum   cum%
  886.65GB 14.06% 14.06%  1019.34GB 16.17%  github.com/ethereum/go-ethereum/trie.(*hasher).hashFullNodeChildren
  472.06GB  7.49% 21.55%   935.83GB 14.84%  github.com/ethereum/go-ethereum/trie.decodeFull
  463.89GB  7.36% 28.90%   463.89GB  7.36%  github.com/ethereum/go-ethereum/trie.decodeRef
     320GB  5.07% 33.98%      320GB  5.07%  github.com/ethereum/go-ethereum/core/vm.(*Memory).Resize
  298.84GB  4.74% 38.72%   298.84GB  4.74%  github.com/ethereum/go-ethereum/rlp.(*encBuffer).makeBytes
  227.90GB  3.61% 42.33%   227.90GB  3.61%  github.com/ethereum/go-ethereum/trie.(*tracer).onRead
  147.57GB  2.34% 44.67%   147.57GB  2.34%  github.com/ethereum/go-ethereum/common.RightPadBytes
  109.50GB  1.74% 46.41%   110.78GB  1.76%  github.com/ethereum/go-ethereum/rlp.(*encBuffer).writeBytes
  106.61GB  1.69% 48.10%   106.61GB  1.69%  golang.org/x/crypto/sha3.NewLegacyKeccak256
  106.55GB  1.69% 49.79%   106.55GB  1.69%  github.com/ethereum/go-ethereum/core/vm.codeBitmap

The main memory allocator is trie loader and trie hasher