Incorrect calculation of score for narrow spectra bins.

[AnthonyOfSeattle]

Normally, spectra will use 100 m/z bins. However, results on synthetic peptide datasets suggest that narrowing that bin size may help for high res data. This needs to be supported at the score level.

https://github.com/AnthonyOfSeattle/pyAscore/blob/25263db3d95fa1b7bfe8c80dd051c78a6ab012e4/pyascore/ptm_scoring/cpp/Ascore.cpp#L33