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
the results below:

```shell
$ 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 celestiaorg#43