Demo data should pull alignment and vcf stats from real-time data, not static hard-coded values.

[tonydisera]

[mvelinder]

Probably not worth it in my opinion. We could always do it later when there's less crunch and less chance of breaking things when people are trying it out for the first time

[tonydisera]

We will need the CRAM for sibling 12877 in order to switch clin.iobio demo (for a quad) to GRCh38. I'm leaving at GRCh37 right now, but I had to make a number of changes to remove the hard-coded QC data. For now, I will rename this issue to 'remove hard-coded QC data for demo' in clin