fix: stac-validate-collections retries should be managed by tpl-at-stac-validate only TDE-1178

[paulfouquet]

Motivation

In the case of a task in tpl-at-stac-validate fails more than twice and the workflow came back failing, the entire stac-validate-collections is retried twice which cause to run the inner stac-validate workflow.

Modification

Avoid tasks in the stac-validate-parallel workflow to retry by default.
Force aws-list-collections to retry twice if it fails.
This is a work around as I could not find a way to specify the retryStrategy at a task level using a templateRef

Checklist

If not applicable, provide explanation of why.

• Tests updated
• Docs updated
• Issue linked in Title

[l0b0]

LGTM

PS: limit is defined as int or string, so we could use integers here.

[paulfouquet]

LGTM

PS: limit is defined as int or string, so we could use integers here.

Weirdly I've got this warning when I use a integer:
Incorrect type. Expected "string".yaml-schema: https://raw.githubusercontent.com/argoproj/argo-workflows/v3.5.5/api/jsonschema/schema.json

[l0b0]

Weirdly I've got this warning when I use a integer: Incorrect type. Expected "string".yaml-schema: https://raw.githubusercontent.com/argoproj/argo-workflows/v3.5.5/api/jsonschema/schema.json

    "io.k8s.apimachinery.pkg.util.intstr.IntOrString": {
      "type": "string"
    },

🤦