Summarize insertions

[donkirkby]

Group identical insertions, and report them along with their prevalance as a percentage of coverage at that location.

• Don't report low quality insertions (<Q30).
• Don't report insertions below 1% prevalance.
• Don't report insertions outside of coordinate regions.
• Combine conseq_ins.csv and coord_ins.csv into a single file.
• Report insertions from G2P alignment.
• Should we translate insertions into amino acid sequences? We could use the surrounding reading frame, but some insertions are not multiples of three nucleotides or are not inserted at codon boundaries. For now, we have decided to go with the nucleotide sequence.