Fix ModelOpt quantized layers startup log

[mmangkad]

Summary

Put back the quantized_layers tuple check that #18182 removed. Without it, ModelOpt mixed-precision checkpoints hit this online-quantization log, try to unpack a dict, and crash at startup. #27284 fixed the same problem once before after the AMD MXFP4 logging was added in #18005.

---

CI States

Latest PR Test (Base): ❌ Run #27499037722
Latest PR Test (Extra): ❌ Run #27499037617

[gemini-code-assist]

Code Review

This pull request simplifies the logic for checking and unpacking quantized_layers during model loading in model_runner.py. It replaces multiple hasattr checks with a direct type and length check on the pre-extracted quantized_layers variable. No review comments were provided, and there is no additional feedback to provide.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[mmangkad]

/rerun-test test/registered/4-gpu-models/test_nvidia_nemotron_3_super_nvfp4.py

[github-actions]

Results for /rerun-test test/registered/4-gpu-models/test_nvidia_nemotron_3_super_nvfp4.py:

🚀 4-gpu-b200 (1 test): ✅ View workflow run

cd test/ && python3 registered/4-gpu-models/test_nvidia_nemotron_3_super_nvfp4.py

[HaiShaw]

#18182 is reverted, we will review with your change, and submit again. Thanks.