Add dynamic retries with increased resource allocation

[adamjtaylor]

I see test runs on tower failing with a 127 exit code indicating out of memory. I suggest adding a tower specific confg profile that sets resource allocations and retry parameters. See example here: https://github.com/Sage-Bionetworks-Workflows/nf-artist/blob/main/nextflow.config