Sequence hashing: Bugfix and slight simplification

[jakobnissen]

Slightly simplify hashing code and fix a minor bug:

• Replace manual bitrotate code with Base.bitrotate (from Julia 1.5)
• Fix an error from a typo in in murmur2 function
• Slightly simplify the code
• Improve test coverage

This results in a ~12% speed increase on my computer for large sequences, but 9% decrease for short sequences. Might just be measurement noise.

[codecov[bot]]

Codecov Report

Merging #229 (607bad0) into master (26e05f7) will increase coverage by 0.52%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #229      +/-   ##
==========================================
+ Coverage   88.73%   89.26%   +0.52%     
==========================================
  Files          31       31              
  Lines        2424     2422       -2     
==========================================
+ Hits         2151     2162      +11     
+ Misses        273      260      -13     

Flag	Coverage Δ
unittests	89.26% <100.00%> (+0.52%)	:arrow_up:

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/longsequences/hash.jl	100.00% <100.00%> (+11.21%)	:arrow_up:
src/bit-manipulation/bitpar-compiler.jl	70.83% <0.00%> (+0.83%)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 26e05f7...607bad0. Read the comment docs.

[jakobnissen]

Merging this, as it's a low-risk change with no user-facing changes, other than the precise hash value has changed, which is fair game considering it's a bugfix.