Upgrade Terraform AWS provider from ~> 5.25.0 to v6

[davidsmejia]

Context

The AWS provider version constraint in infrastructure/backend.tf and infrastructure/batch/versions.tf is pinned to ~> 5.25.0. This pin was added as a hotfix after a staging deploy failed when Terraform resolved to the latest available version (v6.43.0), which introduced breaking changes incompatible with our current configs. The pin keeps deploys working but leaves us on an older provider series and blocks access to newer AWS resource features.

Problem or idea

The following Terraform resources use arguments that were removed or renamed in the AWS provider v6.0.0 release:

• infrastructure/networking.tf — aws_eip.data_refinery_api_ip uses vpc = true (removed; replaced by domain = "vpc")
• infrastructure/batch/compute.tf — all three aws_batch_compute_environment resources use compute_environment_name (renamed to name)
• infrastructure/batch/job_queue.tf — all three aws_batch_job_queue resources use compute_environments = [...] (replaced by a compute_environment_order { } block)

Until these are updated the version pin cannot be bumped.

Solution or next step

1. Update the three files above to use the v6 argument names/shapes (diffs are straightforward — see the v6 upgrade guide)
2. Bump the constraint to ~> 6.0 in both backend.tf and batch/versions.tf
3. Run terraform init -upgrade on the deploy box to pull v6
4. Validate with terraform plan against staging before tagging a release