Write wrapper to build MC3 reference directory

[kamichiotti]

• Address writing permissions problem with /mc3_dev/mc3/util/bgzip.cwl and ./mc3_dev/mc3/util/dbsnp_index/dbsnp_index.cwl, then include in wrapper (as noted in separate Github issue)
• Write *gzi file for human reference
• Write *tbi files for the cosmic and dbsnp references
• Write *fai file for Homo_sapiens_assembly19.fasta
• Temporary addition of Broad Indelocator files from broad_mc3_vcfs.tar.gz for testing purposes.
• Files to pull in:
  1000G_phase1.indels.hg19.sites.vcf
  1000G_phase1.indels.hg19.sites.vcf.idx
  b37_cosmic_v54_120711.vcf.gz
  centromere_hg19.bed
  dbsnp_132_b37.leftAligned.vcf.gz
  Homo_sapiens_assembly19.dict
  Homo_sapiens_assembly19.fasta
  Mills_and_1000G_gold_standard.indels.hg19.sites.vcf
  Mills_and_1000G_gold_standard.indels.hg19.sites.vcf.idx

[kamichiotti]

Move prep_ref subworkflow in mc3_variant.cwl to the reference directory workflow (better to store two references than decompressing reference for each sample).

[kamichiotti]

Include download and preparation of homo_sapiens_vep_86_GRCh37.tar.gz, ExAC_nonTCGA.r0.3.1.sites.vep.vcf.gz and dbNSFPv2.9.1.zip for VEP.

• ftp://ftp.ensembl.org/pub/release-86/variation/VEP/homo_sapiens_vep_86_GRCh37.tar.gz
• ftp://ftp.broadinstitute.org/pub/ExAC_release/release0.3/ExAC.r0.3.sites.vep.vcf.gz
• ftp://ftp.broadinstitute.org/pub/ExAC_release/release0.3/ExAC.r0.3.sites.vep.vcf.gz.tbi
• ftp://dbnsfp:dbnsfp@dbnsfp.softgenetics.com/dbNSFPv2.9.1.zip
  See https://store.docker.com/community/images/opengenomics/variant-effect-predictor-tool for direction