Add GPU E2E stage to Linux VHD builder pipeline

[ganeshkumarashok]

Summary

• Adds a gpu_e2e stage to the Linux VHD builder pipeline (.vsts-vhd-builder.yaml) that runs GPU E2E tests after VHD build completes
• With standalone GPU E2E pipeline triggers disabled in fix: disable automatic e2e pipeline triggers #8135, this ensures GPU E2E tests continue to run as part of the VHD build flow
• The new stage runs in parallel with the existing e2e and scriptless_cse_cmd_e2e stages, all depending on the build stage

Test plan

• Verify the VHD builder pipeline runs the new gpu_e2e stage after build succeeds
• Confirm GPU-tagged E2E scenarios execute with the freshly built VHDs
• Confirm non-GPU E2E stages are unaffected

[github-actions]

PR Title Lint Failed ❌

Current Title: Add GPU E2E stage to Linux VHD builder pipeline

Your PR title doesn't follow the expected format. Please update your PR title to follow one of these patterns:

Conventional Commits Format:

• feat: add new feature - for new features
• fix: resolve bug in component - for bug fixes
• docs: update README - for documentation changes
• refactor: improve code structure - for refactoring
• test: add unit tests - for test additions
• chore: remove dead code - for maintenance tasks
• chore(deps): update dependencies - for updating dependencies
• ci: update build pipeline - for CI/CD changes

Guidelines:

• Use lowercase for the type and description
• Keep the description concise but descriptive
• Use imperative mood (e.g., "add" not "adds" or "added")
• Don't end with a period

Examples:

• ✅ feat(windows): add secure TLS bootstrapping for Windows nodes
• ✅ fix: resolve kubelet certificate rotation issue
• ✅ docs: update installation guide
• ❌ Added new feature
• ❌ Fix bug.
• ❌ Update docs

Please update your PR title and the lint check will run again automatically.

[Copilot]

Pull request overview

Adds GPU E2E coverage back into the Linux VHD build pipeline by introducing a dedicated gpu_e2e stage that runs after the build stage, in parallel with existing E2E stages—ensuring GPU scenarios continue to execute even with standalone GPU E2E pipeline triggers disabled.

Changes:

• Add a new gpu_e2e stage to .pipelines/.vsts-vhd-builder.yaml.
• Configure the stage to run GPU-tagged Linux scenarios (TAGS_TO_RUN=gpu=true) with a custom timeout and capacity-skip behavior.
• Run the stage in parallel with existing e2e and scriptless_cse_cmd_e2e stages (all depend on build).

[Copilot]

Adding gpu_e2e makes a 3rd parallel stage using templates/e2e-template.yaml. That template publishes an artifact named $(LOGGING_DIR), and .pipelines/scripts/e2e_run.sh sets LOGGING_DIR using date +%s (seconds resolution). Parallel jobs that start within the same second can end up with the same artifact name and cause intermittent pipeline failures when publishing artifacts. Consider making the log/artifact name deterministically unique per job/stage (e.g., include stage/job ID or use higher-resolution time) so concurrent E2E stages can’t collide.

[awesomenix]

Lets pick some few cases of GPU to run, tag them with GPU_Basic or something like that and run it part of PR