Feature request: extract umi's from header in fastqToBam step

Description of feature

fgbio's FastqToBam supports the option --extract-umis-from-read-names true/ -n true to extract umis from the readnames. Although the pipeline is at the moment not able to receive the extra config info in the samplesheet. This both can be solved by autodetection of the said parameter before running FastqToBam.

My suggestion is something like this:

args_to_append=''
if python3 -c 'import sys,re;sys.exit(not re.fullmatch(r"^[ATCGN]+(\+[ATCGN]+)?$",open(sys.argv[1]).readline().split()[-2].split(":")[-1]))' <(zcat example.fastq.gz|head -n 1); then
    args_to_append='--extract-umis-from-read-names true'
fi

This should support current illumina format and maybe also sra format.

Background info

Illuminas spec for fastq files allows encoding of UMI sequences in the header (using wikipedia fastq record header example updated for current illumina spec):

@EAS139:136:FC706VJ:2:2104:15343:197393:AATCG+AATCG 1:Y:18:ATCACG
@SEQUENCER:RUN:FLOWCELL:LANE:COORDINATE:COORDINATE:COORDINATE:UMI+UMI PAIR:PF:EVENNUMBER:INDEX

Start by locating the fastqToBam step and the samplesheet configuration path. Inspect how the first read from an example.fastq.gz header is parsed, including the current Illumina format and possible SRA format, then verify that the completed pipeline detects header UMIs and passes fgbio's extraction option to FastqToBam.