Add signalMatch start and stop to output

[tijeco]

This is kind of related to #27. Basically, the signal HMM searches the entire protein, ideally the hit is at the beginning, but it would be nice to know for sure where the match was. Sometimes prodigal may overextend the 3' start codon and tack on a few more residues/ pick the wrong methionine. Now, it is hard to say to what extent these artifacts are correct. Who knows, maybe in the proteome it actually happens that way. I just know that when looking at a multiple seuence alignment it is very clear that the 3' end of some of the hits are a bit longer than usual.

[tijeco]

I'll have to put this on hold, there doesn't seem to be an obvious output from pyhmmer that includes this. I thought this was in hmmer output?? I;l just have to spend more time on this.