Modlin ibs

[biominer]

Added the bed file and edited region_filter.py and the README.md accordingly to add and document the option for masking variants in the regions of systematic low coverage/poor mappability reported by Modlin et al. In response to #2 .

[pvanheus]

@biominer I'm finally getting around to reviewing this PR. The Modlin et al paper links to a CSV file on Zenodo but the regions seem to differ from the BED file that you provided. Here is the start of the BED file I computed from the Zenodo CSV:

NC_000962.3 0 33
NC_000962.3 60 61
NC_000962.3 1589 1591
NC_000962.3 1592 1595

(note that this is BED, so coordinates are zero-based, half-open)