Description

The set-up scripts for Helm chart testing (e.g., set-up-test.sh, set-up-multi-dedicated-test.sh, set-up-multi-shared-test.sh) currently don't provide an easy mechanism to load locally-built Docker images into the kind cluster for debugging purposes.

Proposed Solution

Add support for automatically detecting and loading locally-built clp-package images from build/clp-package-image.id into the kind cluster. This functionality should:

1. Check if a locally-built image exists at build/clp-package-image.id
2. Generate an appropriate dev tag (e.g., dev-{user}-{short-hash})
3. Load the image into the kind cluster using kind load docker-image
4. Configure Helm to use the local image with pullPolicy=Never

Reference Implementation

A working example for set-up-test.sh is provided in the PR comment showing the pattern to follow.

Context

This feature request was raised during PR review to improve developer experience when testing Helm chart changes with locally-built images.

Related Links

• PR: feat(helm): Add multi-node deployment support with configurable scheduling. #1829
• Comment: feat(helm): Add multi-node deployment support with configurable scheduling. #1829 (comment)
• Requested by: @junhaoliao