Optimization of slices sums (NEON Instructions)

[AndreyyTs]

Optimization of Slice Addition for ARM64

Tests have passed.

Acceleration results are below:

Benchmark Name	Iterations	Time per Operation (ns/op)
BenchmarkSumUnit8	40464	29861
BenchmarkSumUnit8Generic	2793	395114
BenchmarkSumUnit16	19112	62727
BenchmarkSumUnit16Generic	2646	411772
BenchmarkSumUnit32	9080	128728
BenchmarkSumUnit32Generic	2186	460634
BenchmarkSumUnit64	4209	279131
BenchmarkSumUnit64Generic	1989	505266

[seg-atlantis-prod[bot]]

Atlantis commands can't be run on fork pull requests. To enable, set --allow-fork-prs or, to disable this message, set --silence-fork-pr-errors

[seg-atlantis-prod[bot]]

Error parsing command: EOF found when expecting closing quote

[AndreyyTs]

@kalamay Hello Can you take a look at the pull request?