fgbio_create_umi_consensus: take aligner index instead of fasta

[matthdsm]

Is your feature request related to a problem? Please describe

The subworkflow shouldn't take a fasta file and create an aligner index - out of scope xref #2302 xref #2179 xref #2185

Describe the solution you'd like

Pregenerate the aligner index and use that as input. Doing this for each sample creates a huge amount of overhead

Describe alternatives you've considered

No response

Additional context

No response