Migrate all resources to S3 bucket

[kelly-sovacool]

This will allow us to get rid of the FRCE & Biowulf-specific code as well as redundant resources. We can also add a command to locally download the resources.

RENEE/config/genomes/biowulf/hg38_30.json

Lines 20 to 22 in 5fb4d14

	"FUSIONBLACKLIST": "s3://nciccbr/Resources/RNA-seq/arriba/blacklist_hg38_GRCh38_v2.0.0.tsv.gz",
	"FUSIONCYTOBAND": "s3://nciccbr/Resources/RNA-seq/arriba/cytobands_hg38_GRCh38_v2.0.0.tsv",
	"FUSIONPROTDOMAIN": "s3://nciccbr/Resources/RNA-seq/arriba/protein_domains_hg38_GRCh38_v2.0.0.gff3"

We should also split up genome config files into species vs annotation specific files to avoid file duplication.

May want to wait on this until we migrate to nextflow, which has better design patterns for genome configs.