feat: add infra filters for terraform, nix, ansible, and opentofu

[BSD-Yassin]

Summary

• Add first-class rtk terraform, rtk nix, and rtk ansible-playbook support for compact infra output with gain tracking.
• Add first-class rtk tofu support as a Terraform-equivalent command by reusing the Terraform filter pipeline and preserving OpenTofu-specific tracking labels.
• Extend discover rewrite coverage so terraform, nix, ansible-playbook, and tofu commands are recognized and rewritten to RTK equivalents.

Test plan

• cargo fmt --all
• cargo clippy --all-targets
• cargo test --all
• Runtime validation in nix-shell with real tools (terraform, opentofu, ansible, nix) using raw-vs-rtk comparisons:
  • Terraform plan: 20 -> 3 lines
  • OpenTofu plan: 20 -> 2 lines
  • Ansible playbook: 11 -> 4 lines
  • Nix search: 113383 -> 21 lines
• Verify gain tracking entries for all commands:
  • rtk terraform plan -no-color
  • rtk tofu plan -no-color
  • rtk nix search nixpkgs hello
  • rtk ansible-playbook ...
• Validate discover rewrite behavior:
  • rtk rewrite "tofu plan -lock=false"
  • rtk rewrite "terraform apply -auto-approve"
  • rtk rewrite "nix search nixpkgs hello"
  • rtk rewrite "ansible-playbook ..."

Notes

• Smoke script run (bash scripts/test-all.sh) reports known environment-related failures unrelated to this change (learn/discover need Claude sessions, and script used an older globally installed rtk binary).

[CLAassistant]

All committers have signed the CLA.

[pszymkowiak]

[w] wshm · Automated triage by AI

📊 Automated PR Analysis

✨ Type	feature
🟡 Risk	medium

Summary

Adds first-class RTK filter support for Terraform, OpenTofu, Nix, and Ansible-playbook commands, providing compact output with gain tracking. Also extends the discover/rewrite system to recognize and rewrite these commands to their RTK equivalents.

Review Checklist

• Tests present
• Breaking change
• Docs updated

---

Analyzed automatically by wshm · This is an automated analysis, not a human review.

[BSD-Yassin]

Those tools are heavy in my daily workflow, and while debugging an issue, they can be launched in conjunction with aws-cli, docker, curl and eat a lot of tokens. Take this PR as my first reduction of token and the start of a lot of "infra/devops" related reduction.

Note : I have not added any automated tests. That could obviously be a blocking step that can be discussed here.

The nix command should be alright to test directly on the pipeline or locally but for terraform, ansible and opentofu, I didn't want to create a whole infra for testing each use case.

Nor did I want to create an output model file, as I assume that any of those tools will evolve in the future and it would require this output file would need to follow version changes. I have yet to think of a fair solution, I suppose all of them are able to create local resources, that would make a good compromise for an execution test all while not reaching a remote instance to provision / configure.

Feel free to ask any questions, remarks or require any improvements, as it is one of my rare contributions, and do note that I obviously helped myself with IA to produce it.