Add support for aligning untrimmed reads

[dfornika]

There may be cases where we'd like to use the raw, untrimmed reads for alignment. For example, if we want to get the most accurate estimate of the sequencing error rate then we would be excluding some low-quality sequence if we were to use the trimmed reads for the alignment.

Add an --align_untrimmed_reads flag that will cause the pipeline to use the untrimmed reads for alignment. This is also supported in the BCCDC-PHL/amplicon-consensus pipeline, for similar reasons.