[WORKFLOWS-70] Create public buckets for iGenomes reference files

[BrunoGrandePhD]

I was hoping to re-use the s3-bucket-v2.j2 template from aws-infra, but I didn't want to have to deal with the S3objects macro, especially because it's not relevant to our use case. I opted for adapting the template for our purposes.

I'm still figuring out how to copy the files. Unfortunately, interacting with the nf-core bucket requires using --no-sign-request, which then prevents the requests to our bucket to work since they're unauthenticated (see example command below). I'm currently staging files locally on a Service Catalog instance. I included a script that I will run once the bucket is deployed to production.

aws s3 --no-sign-request sync "s3://source-bucket/" "s3://target-bucket/"

[BrunoGrandePhD]

I can confirm that the region restrictions works as expected:

# From a Service Catalog instance in us-east-1
$ aws s3 cp s3://sage-igenomes-dev/igenomes/Homo_sapiens/Ensembl/GRCh37/Sequence/Chromosomes/MT.fa - | head -n 1
>MT

# From my personal computer in Seattle
$ aws s3 cp s3://sage-igenomes-dev/igenomes/Homo_sapiens/Ensembl/GRCh37/Sequence/Chromosomes/MT.fa - | head -n 1
download failed: s3://sage-igenomes-dev/igenomes/Homo_sapiens/Ensembl/GRCh37/Sequence/Chromosomes/MT.fa to - An error occurred (403) when calling the HeadObject operation: Forbidden

[thomasyu888]

Lgtm