feature: Refine Functionality Test Scope to Focus on Router Image and Helm Chart

Describe the feature

The current functionality test in https://github.com/vllm-project/production-stack/blob/main/.github/workflows/functionality-helm-chart.yml appears to be testing more than its intended scope. While the primary purpose should be verifying the router image build and Helm chart deployment (as seen in the Docker build command docker build --build-arg INSTALL_OPTIONAL_DEP=default -t localhost:5000/git-act-router -f docker/Dockerfile .) in link it currently includes testing of vllm/vllm-openai image deployments through values files like (values-04-multiple-models.yaml)[https://github.com/vllm-project/production-stack/blob/24049186d02329fb52ef02a4b91846a6afb46150/.github/values-04-multiple-models.yaml#L5].

Expected Behavior:
The functionality test should focus exclusively on:

1. Validating the router image build process
2. Testing the Helm chart deployment for the router component
3. Confirming basic router functionality

Suggested Changes:

1. Remove vllm/vllm-openai specific tests from the functionality workflow
2. Add explicit test cases for router-specific scenarios
3. Consider separating vllm/vllm-openai tests into a different workflow if needed

This change would better align the test scope with the actual purpose of this workflow and improve test clarity.

Reference Workflow Example

• https://github.com/juicedata/juicefs/blob/main/.github/workflows/chaos.yml
• https://github.com/apache/dolphinscheduler/blob/dev/.github/workflows/e2e-k8s.yml

Why do you need this feature?

No response

Additional context

No response

Start by reading .github/workflows/functionality-helm-chart.yml, especially the Docker build command, and inspect .github/values-04-multiple-models.yaml to identify vllm/vllm-openai coverage. Done means the workflow validates the router image build, router Helm deployment, and basic router functionality, with vllm/vllm-openai tests removed or separated.