Failed to report errors in plan

[mdtro]

On this PR: https://github.com/getsentry/ops/pull/10827 errors from the plan were not reported via CI. @mwarkentin was able to run the plan locally and forward me the errors: https://github.com/getsentry/ops/pull/10827#issuecomment-2148672192

Here's a link to the relevant actions run: https://github.com/getsentry/ops/actions/runs/9376279383

[mwarkentin]

https://github.com/getsentry/ops/pull/11045#issuecomment-2178869420

https://github.com/getsentry/ops/actions/runs/9583980312

Had to run locally to get:

│ Error: Invalid value for input variable
│
│   on main.tf line 128, in module "instances":
│  128:   additional_disks = [
│  129:     google_compute_disk.data[each.key].name,
│  130:   ]
│
│ The given value is not suitable for module.instances.var.additional_disks
│ declared at
│ .terraform/modules/instances/instance_salt/_variables.tf:105,1-28: map of
│ object required.
╵
ERRO[0014] Terraform invocation failed in /Users/mwarkentin/repos/sentry/ops/terragrunt/regions/multi-tenant/clickhouse/snuba-generic-metrics-distributions/de/.terragrunt-cache/hoGDh_7488FOX2UZWs4a6aaiLiM/0rghWROp0gJdHK4BUmBouoAzp4g/cluster_clickhouse  prefix=[/Users/mwarkentin/repos/sentry/ops/terragrunt/regions/multi-tenant/clickhouse/snuba-generic-metrics-distributions/de]
ERRO[0014] 1 error occurred:
	* exit status 1

[mwarkentin]

I was able to get this plan working without any further changes locally by removing .terragrunt-cache/, not sure why this isn't working on GHA, I thought it inits from scratch each time.

[Dav1dde]

I have a few failing plans here https://github.com/getsentry/ops/actions/runs/9696824394/attempts/1#summary-26759629510 but the output has no explanations on what went wrong.