Commitment Experiments

inherits experimental tasks from https://github.com/erigontech/erigon/issues/13884

# Experiments

- [ ] https://github.com/erigontech/erigon/issues/17838
- [x] Binary Patricia Trie revive: add missing bits and adapt binary trie optionally (with commitment rebuilding possibility) -- #21791
- [ ] ~~memo: need to research if we can use same field cell.hash to keep memoised state hashes (which essentially state roots of those cells). During my initial work, i noticed that cell.hash and cell.stateHash never matches but we keep 2 fields for each cell having same meaning but memo hash is safe to use while cell hash may be even some intermediate value.~~ cell.hash is not empty when its branch, otherwise its not written. having that fact, need to check that we DO not write stateHash when cell.hash is not empty (meaningless) For cells, cell.hash is empty. just double check that we have correct understanding.
- [x] rebuild commitment for some chains without key referencing https://github.com/erigontech/erigon/issues/20010
- [ ] make witness and proof gathering during unfolding/folding the trie - should simplify proof support and speed it up
- [x] https://github.com/erigontech/erigon/issues/20586