test: Add L0_pytorch_python_runtime coverage for default_model_filename autofill

[mattwittwer]

What does the PR do?

Related to to triton-inference-server/core#518. Strengthens L0_pytorch_python_runtime so the core autofill regression fails with a direct diagnosis instead of an opaque constants.pkl load error:

• test.sh: fail if neuralnet takes the TorchScript load path, and assert the served model config reports an empty default_model_filename.
• unit_test.py: three _get_model_path cases — model.py preferred over a sibling weights model.pt, an explicit default_model_filename is honored, and a TorchScript-only model still resolves to model.pt.

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:

Check the conventional commit type
box here and add the label to the github PR.

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

Related PRs:

triton-inference-server/core#518

Where should the reviewer start?

Test plan:

• CI Pipeline ID:

60326198

Caveats:

Background

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

• closes GitHub issue: #xxx

[greptile-apps]

Greptile Summary

Adds regression coverage for Python-runtime model filename selection.

• Validates that the served neuralnet configuration leaves default_model_filename empty and reports retrieval or parsing failures.
• Detects accidental selection of the TorchScript loading path.
• Adds unit tests covering Python-model preference, explicit filename selection, and TorchScript-only fallback.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains.

Important Files Changed

Filename	Overview
qa/L0_pytorch_python_runtime/test.sh	Adds an end-to-end regression guard and now correctly treats configuration retrieval, parsing, and model-name validation failures as test failures.
qa/L0_pytorch_python_runtime/unit_test.py	Adds isolated temporary-directory tests for the three intended model-path resolution cases.

_{Reviews (3): Last reviewed commit: "update failure case handling" | Re-trigger Greptile}

[Vinya567]

LGTM