Properly sort chromosomes in output file

[wdecoster]

Sorting the output file should be done by chromosome and then by position. However, sorting by chromosome is hard: you don't want chr10 before chr7, chromosomes may or may not have chr prefix. Removing the 'chr' and converting to int is a problem for chrX, chrY, chrMT, and all decoy contigs/alternative haplotypes.

I naively tried key=lambda col: col.astype(str).str.replace('chr', '').astype(int)) but it has to be more advanced.

Alternatively, we could use bcftools... but adding more dependencies if something I would like to avoid. Then again I already added pandas for combining the files.