Feature/DIMS_HMDBparts_bymz

[mraves2]

HMDB database (including adducts and isotopes) is divided into parts to enable parallel processing. In the old setup, each part contained 20000 lines; in the new set-up, each part contains a certain mass range. For m/z < 100, smaller ranges are used because of the higher abundance of peaks at lower masses.
There are no modifications in the DIMS repo necessary for this change.

[rernst]

One minor comment, let me know if you would like to update the code.

[rernst]

To separate some logic (prepare for tests and easier to maintain) I would write this like:

while (segment_end < max_mz) {
    if (segment_start < 100) {
      segment_size = 5
    } else {
      segment_size = 10
    }
    mz_segments <- c(mz_segments, segment_start)
    segment_start <- segment_start + segment_size
    segment_end <- segment_end + segment_size
}

[mraves2]

Good suggestion; I implemented the modification.