perf(merkle): optimize hashing by reusing sha256 in a pool

[odeke-em]

This change optimizes SHA256 hashing for leafHash & innerHash by pooling and reusing hashes, which massively reduces on RAM consumption and even CPU time (benchmarking on my noisy machine is skewed but the results can be verified individually) producing results below:

$ benchstat before.txt after.txt
name                          old time/op    new time/op    delta
HashAlternatives/recursive-8    77.6µs ± 2%    77.0µs ± 2%     ~     (p=0.165 n=10+10)
HashAlternatives/iterative-8    76.3µs ± 1%    76.2µs ± 3%     ~     (p=0.720 n=9+10)

name                          old alloc/op   new alloc/op   delta
HashAlternatives/recursive-8    25.4kB ± 0%     6.4kB ± 0%  -74.94%  (p=0.000 n=10+10)
HashAlternatives/iterative-8    28.1kB ± 0%     9.1kB ± 0%  -67.78%  (p=0.000 n=10+10)

name                          old allocs/op  new allocs/op  delta
HashAlternatives/recursive-8       497 ± 0%       199 ± 0%  -59.96%  (p=0.000 n=10+10)
HashAlternatives/iterative-8       498 ± 0%       200 ± 0%  -59.84%  (p=0.000 n=10+10)

Fixes #43

Overview

Checklist

• New and updated code has appropriate documentation
• New and updated code has new and/or updated testing
• Required CI checks are passing
• Visual proof for any user facing features like CLI or documentation updates
• Linked issues closed with keywords

[rootulp]

These changes seem fine to me but it's difficult to approve because the modified functions don't have unit tests. They should've had unit tests prior to this PR but I expect they don't because the merkle package is copy + pasted from the CometBFT repo.

IMO we need to decide if it's okay for the merkle package in this repo to diverge from the upstream CometBFT implementation. cc: @cmwaters

[odeke-em]

@rootulp how come the code was copied and pasted in here without the appropriate attributions/licenses or without just plain imports or submodules?

[rootulp]

I think this is relevant:

go-square/merkle/README.md

Line 6 in 4e84d80

This is a forked copy of github.com/cometbft/cometbft/crypto/merkle

I suppose we could also copy + paste this license into that directory.

[cmwaters]

Closing this because for now we've decided to keep the merkle library in our cometbft fork. Thanks for opening the PR in any case. Sorry that we didn't merge it. We may look to apply this to our cometbft fork in the future