✨ Add llama_stack to release workflows

[fabianvf]

Summary by CodeRabbit

Release Notes

• New Features
  • Added support for configurable llama-stack image across bundle generation and deployment
  • Llama-stack image reference can now be customized through configuration parameters with sensible defaults
  • Introduced environment variable support for flexible image reference management during deployment

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

Adds llama_stack image parameter propagation across the build pipeline and deployment configuration, including GitHub Actions inputs, workflow bundle arguments, Kubernetes manifests, Helm charts, and Ansible role defaults.

Changes

Cohort / File(s)	Summary
GitHub Actions & Workflows .github/actions/make-bundle/action.yml, .github/workflows/create-release.yml	Added llama_stack input to the make-bundle action and propagated it through bundle generation commands with conditional option appending. Updated release workflow to include llama_stack image in bundle build step arguments.
Kubernetes & Container Configuration bundle/manifests/konveyor-operator.clusterserviceversion.yaml	Added llama-stack to relatedImages section and introduced RELATED_IMAGE_LLAMA_STACK environment variable in the tackle-operator container.
Helm Configuration helm/values.yaml, helm/templates/deployment.yaml	Added llama_stack image entry with default value in values.yaml and corresponding RELATED_IMAGE_LLAMA_STACK environment variable in deployment template.
Ansible Configuration roles/tackle/defaults/main.yml	Converted kai_llm_proxy_image_fqin from static image reference to environment-driven lookup with fallback default.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~15 minutes

• The changes follow a consistent, repetitive pattern of adding the same image reference across different configuration layers, which reduces cognitive load per file
• However, the spread across 6 files in different formats (GitHub Actions YAML, Kubernetes manifests, Helm templates, Ansible defaults) requires familiarity with multiple configuration styles
• The environment variable indirection in roles/tackle/defaults/main.yml warrants verification that the Jinja2 lookup syntax is correct and the fallback behavior is as intended

Possibly related PRs

• 🌱 Add kantra image to bundle #464: Makes identical configuration propagation pattern for a different image name, demonstrating consistency in this approach
• 🐛 add kai to make-bundle action #468: Directly related modification to the make-bundle action adding optional image inputs and conditional option appending

Suggested reviewers

• djzager

Poem

🐰 A new stack arrives, from llama's embrace,
Through workflows and Helm, it finds its place,
From actions to manifests, configs align,
Each layer now knows the image to assign!
The operator stands ready, no stone left unturned.

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title mentions adding llama_stack to release workflows, which aligns with the changes across GitHub Actions, Helm, and Ansible configurations that introduce llama_stack image support to the release process.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between fc1083f and 96f65ff.

📒 Files selected for processing (6)

• .github/actions/make-bundle/action.yml (2 hunks)
• .github/workflows/create-release.yml (1 hunks)
• bundle/manifests/konveyor-operator.clusterserviceversion.yaml (2 hunks)
• helm/templates/deployment.yaml (1 hunks)
• helm/values.yaml (1 hunks)
• roles/tackle/defaults/main.yml (1 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: run-ci / e2e-api-integration-tests

🔇 Additional comments (7)

helm/templates/deployment.yaml (1)

66-67: LGTM! Environment variable correctly references Helm values.

The RELATED_IMAGE_LLAMA_STACK environment variable is properly configured using the Helm template syntax and follows the same pattern as other related image environment variables.

.github/workflows/create-release.yml (1)

231-231: LGTM! Release workflow correctly includes llama_stack with mirrored image.

The llama_stack parameter uses the quay.io/konveyor registry, which is correct for the release process. The image mirroring step (lines 187-210) will copy external images like docker.io/llamastack/distribution-starter to quay.io/konveyor/distribution-starter during the release.

bundle/manifests/konveyor-operator.clusterserviceversion.yaml (2)

362-363: LGTM! Environment variable properly added to ClusterServiceVersion.

The RELATED_IMAGE_LLAMA_STACK environment variable follows the established pattern for other related images in the operator deployment specification.

---

567-568: LGTM! Related image entry correctly configured.

The llama-stack entry in the relatedImages list uses the same image reference as the environment variable, maintaining consistency. The hyphenated name format matches the convention used for other related images.

.github/actions/make-bundle/action.yml (2)

64-67: LGTM! Input parameter properly defined.

The llama_stack input follows the established convention for other image inputs in this action, with appropriate description, optional requirement, and empty default value.

---

107-107: LGTM! Conditional logic correctly propagates llama_stack to bundle options.

The conditional check and OPTS append follow the same pattern as other image parameters, ensuring the llama_stack value is properly passed to the Helm-based bundle generation process.

roles/tackle/defaults/main.yml (1)

332-332: LGTM! Environment-driven image configuration correctly implemented.

The change to kai_llm_proxy_image_fqin properly reads from the RELATED_IMAGE_LLAMA_STACK environment variable with an appropriate fallback to the default image. This aligns with the pattern used for other image configurations in this file (e.g., hub_image_fqin on line 28) and integrates well with the environment variables added in the deployment manifests.