[mc-nv]

What does the PR do?

• trtexec is no longer in the default Triton container, so QA tests that built plans at run time fail.
• Add gen_qa_image_trt_plans.py: builds resnet50/resnet152/vgg19 plans (TensorRT Python API) from the exported ONNX, in the TensorRT generation stage.
• L0_device_memory_tracker copies the prebuilt plans.

Checklist

• PR title reflects the change and is of format <commit_type>: <Title>
• Changes are described in the pull request.
• Related issues are referenced.
• Populated github labels field
• Added test plan and verified test passes.
• Verified that the PR passes existing CI.
• Verified copyright is correct on all changed files.
• Added succinct git squash message before merging ref.
• All template sections are filled out.
• Optional: Additional screenshots for behavior/output changes with before/after.

Commit Type:

• build
• ci
• docs
• feat
• fix
• perf
• refactor
• revert
• style
• test

Related PRs:

• docs: Note trtexec is not in the default Triton container python_backend#462

Where should the reviewer start?

qa/common/gen_qa_image_trt_plans.py, then its wiring in qa/common/gen_qa_model_repository and gen_qa_model_repository.py.

Test plan:

• Generate the QA model repository; confirm qa_dynamic_batch_image_model_repository/{resnet50,resnet152,vgg19}_plan/1/model.plan exist.

• Run L0_device_memory_tracker in a container without trtexec.

• Not yet run: generator and test are unverified on a GPU.

• CI Pipeline ID: pending

Caveats:

• Plans are GPU-arch and TRT-version specific; generation must run per architecture.
• TensorRT stage is skipped for MODEL_TYPE=igpu.
• L0_memory_growth, L0_perf_resnet, L0_perf_deeprecommender still use trtexec and fail first on a missing perf_model_store; out of scope.

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

• Resolves: TRI-2045