perf(l1): use our keccak implementation for receipt blooms

[MegaRedHand]

Motivation

Receipts computation is ~3% of block execution. It's not much, but it should be easy to optimize, since we didn't touch it much.

Description

This PR changes the receipt bloom computation to use our keccak implementation (before it was using tiny_keccak inside Bloom::accrue). It also adds some preallocation to the receipt encoding, which should account for the smallest transactions, at least.

[github-actions]

Lines of code report

Total lines added: 44
Total lines removed: 0
Total lines changed: 44

Detailed view

+---------------------------------------+-------+------+
| File                                  | Lines | Diff |
+---------------------------------------+-------+------+
| ethrex/crates/common/types/receipt.rs | 387   | +44  |
+---------------------------------------+-------+------+

[Copilot]

Pull request overview

This PR optimizes receipt bloom computation by switching from tiny_keccak (used internally by BloomInput::Raw) to the codebase's custom keccak implementation. The changes also include preallocation optimization for receipt encoding and improved variable naming.

Key Changes

• Switched bloom filter computation to use custom keccak_hash function with BloomInput::Hash instead of BloomInput::Raw
• Added preallocation (512 bytes) to encode_inner_with_bloom to reduce allocations
• Improved variable naming from encode_buff to encode_buf
• Added comprehensive test to verify bloom computation correctness

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review for a chance to win a $100 gift card. Take the survey.

[github-actions]

Benchmark Block Execution Results Comparison Against Main

Command	Mean [s]	Min [s]	Max [s]	Relative
base	61.100 ± 0.248	60.768	61.540	1.00
head	61.329 ± 0.222	60.963	61.713	1.00 ± 0.01