reduced allocations in Bare for big speedup

[guigzzz]

Benchmark running on a 7700HQ laptop

goos: windows
goarch: amd64
BenchmarkDigestHexPacket-8       1000000          1287 ns/op         464 B/op         13 allocs/op
BenchmarkDigestPacket-8          1000000          1192 ns/op         400 B/op         11 allocs/op
BenchmarkBarePacket-8            2000000           920 ns/op         400 B/op         11 allocs/op
BenchmarkDigestHex-8             2000000           767 ns/op         256 B/op          3 allocs/op
BenchmarkDigest-8                2000000           675 ns/op         192 B/op          1 allocs/op
BenchmarkBare-8                  3000000           428 ns/op         192 B/op          1 allocs/op
PASS
coverage: 36.5% of statements

This can still be improved, see TODO in Bare function

[dreadl0ck]

Awesome, great job! Thanks :)

[dreadl0ck]

I've updated the benchmarks and added length checks in case the arrays are empty, to prevent accessing the index -1.

Thanks again for your contribution!