Disable AWS launch_template from nebari-config schema

[viniciusdc]

Reference Issues or PRs

What does this implement/fix?

This PR disables the user-facing object from the nebari-config schema while still keeping the logic introduced initially in #2621.

Motivation

The #2842 attempted to disable the ami_id setting from the launch_template configuration under the node groups. However, this change caused Terraform deployment issues because the aws node_group relied on the ami_id attribute internaly, leading to a failed Nebari deployments when it was removed in the linked PR.

Additionally, we've encountered recent internal connection issues with the nodes using just the pre_bootstrap_command. These issues indicated that the current configuration for launch_template is not functioning correctly.

It also includes:

• a custom error message in the schema validator for letting users know this is a temporary removal;
• A better docstring to the auxiliary construct_aws_ami_type function since the choice for the CUSTOM instance type depends on the presence of both launch_template and ami_id.

Put a x in the boxes that apply

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds a feature)
• Breaking change (fix or feature that would cause existing features not to work as expected)
• Documentation Update
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no API changes)
• Build related changes
• Other (please describe):

Testing

• Did you test the pull request locally?
• Did you add new tests?

How to test this PR?

• init an aws config file
• (optional) run a deployment;
• Include a new node group with the launch_template field. This would simulate a scenario where the user has such a config already (for example):

      launch_template:
        # Replace with your custom AMI ID
        ami_id: ami-0abcdef1234567890
        # Command to run before the node joins the cluster
        pre_bootstrap_command: |
          #!/bin/bash
          # This script is executed before the node is bootstrapped
          # You can use this script to install additional packages or configure the node
          # For example, to install the `htop` package, you can run:
          # sudo apt-get update
          # sudo apt-get install -y htop"

Though, just adding launch_template will trigger the error.

• run nebari render or nebari validate to trigger the load of the schemas. This should lead you to a ValueError

Any other comments?

[viniciusdc]

Tested as follows:

This PR was tested in a previously deployed nebari instance on AWS that had node groups using the launch_template. When attempting the re-deployment from this version, the following (expected) message occurs:

As the launch_template configuration is heavily coupled with the instance type used for both normal nodes and GPU utilization, I also tested to make sure both could be fully utilized:

• Usual node group with GPU showing correctly assigned AWS instance type

    gpu:
      instance: g4dn.xlarge
      min_nodes: 0
      max_nodes: 5
      gpu: true
      single_subnet: false
      permissions_boundary:

[viniciusdc]

I am testing a fresh init and deploy just to be safe.

Update:
As shown above, deployments are working as expected, and the appropriate instance_types are been assigned. I also tested this afternoon a nebair init to double check new deployments will not be affected by the removal of the template schema options.