RFC-0063 live parity was not run #2033

Closing #272, #277, #308 and #2030 without running what they still ask for. The RPC cost is the reason. The code is on #2032.

scripts/rfc0063-gate.sh is the remainder. Phase 1 is vitalik.eth (0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045), Ethereum 18,000,000 to 20,000,000, hash sets against the pinned Etherscan files: txlist 2,318, tokentx 3,334, tokennfttx 106, token1155tx 53, and txlistinternal for the same range with field parity and the same traceId by address and by txhash. Phase 3 is the pinned odd cases in tests/fixtures/rfc0063. Phases 4 and 5 are 0x7a25bd5f286fb722e7578c62e86a675ff0a00b15, blocks 17,034,870 to 17,300,000: 3,131 withdrawals from block bodies, about 269,000 eth_getBlockByNumber calls, the pinned getminedblocks cases, and reward parity to the wei. Phase 7 needs MIRROR: one partition built twice is byte-identical, a broken partition is refused, and getblocknobytime matches Etherscan, pre-London included.

#2030 is the other piece. #2032 skips a_wide_batch_stays_under_the_cursor in the mutants job. The red scheduled run is https://github.com/nuthatch-org/nuthatch/actions/runs/38128606532. The bot refiles that failure once this predecessor is closed, so a green scheduled mutants (delightful core) run is what retires it.