Make it so that filter-reads-alignments.py can also save the corresponding JSON

[terrycojones]

filter-reads-alignments.py (see #528) can filter reads alignments and prints the resulting reads. Make this more useful by adding a --saveJSON (or similar) filename option that it writes the filtered reads' alignment (hits, matches, call it what you will) to. That way we will be able to quickly extract a filtered set of reads and match results from a much larger overall result (I have previously done this by hand).