minor feature request: same strand alignment only

[Magdoll]

This is a minor feature request (not urgent): adding a parameter so that alignments are only done on the same strand.

For PacBio transcriptome data (which often are already stranded), there is no need to look for reverse complement matches.

Not sure how much time / memory this saves, so just putting this request out there.

Thanks,
--Liz

[pb-cdunn]

I think this should save 1/2 time and 1/2 memory, right? Also, 1/2 disk-space.

[pb-jchin]

the time save may not be 1/2. The indexing part is typically faster than the alignment confirmation. As there is no hit on the RC part, we don't save the alignment confirmation. However, it makes the semantics of an anlysis process much more clean. In some case, we don't need RC alignment. It is a good thing.