Migrate KEDA Autoscaling from CloudWatch Metrics to SQS Queue Depth

[magniloquency]

Description

This PR migrates KEDA from scaling based on CloudWatch metrics to scaling directly based on SQS queue depth and closes #94.

I modified all of the files listed in the original issue, minus helm.tf. I analyzed this file and looked at the helm charts it is releasing and did not think any changes were necessary. Please let me know if I missed something.

I updated the IAM permissions for SQS and restricted them to only the queue that KEDA needs, following the principle of least privilege.

I updated documentation where necessary. I think 40_submitting_multiple_tasks may also need an update, but I am not sure.

I left the metrics lambda in place because I am not sure if it has other uses. Let me know if we want to remove it.

Note: I was not able to deploy and test this myself.

Note: We are using the identityOwner attribute in hpa.yaml. KEDA has deprecated this as of v2.13 and plans to remove it in v3. I have kept this as-is but it may be worth revisiting in the future.

Checklist

• Added tests that cover your change (if possible)
• Added/modified documentation as required (such as the README.md, or the docs directory)
• Manually tested
• Made sure the title of the PR is a good description that can go into the release notes
• (Core team) Added labels for change area (e.g. area/controlplane) and kind (e.g. kind/improvement)

BONUS POINTS checklist:

• Backfilled missing tests for code in same general area
• Refactored something and made the world a better place