[https://nvbugs/6428063][chore] Unwaive the DeepSeekV3Lite nvfp4 pp4 test

[lori-ren]

Description

use_host_stop_criteria (added by #15920) was reverted wholesale by #16163, so the mechanism
behind nvbugs/6428063 no longer exists — grep -rn use_host_stop_criteria returns nothing on
main. The waiver has been skipping this case since 2026-07-08 and has not been touched since.

This removes that one line and nothing else:

$ git diff --stat
 tests/integration/test_lists/waives.txt | 1 -

Supersedes #16136, whose fix (threading the flag through the PP ring-broadcast payload in
py_executor.py) targets code that no longer exists: against today's main its getattr
always yields False and its hasattr guard never fires.

Test Coverage

The un-waived case itself:

accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=0-pp4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-low_precision_combine=False-torch_compile=True]

It is listed in l0_dgx_b300.yml, l0_gb300_multi_gpus.yml and l0_rtx_pro_6000.yml, all
stage: post_merge. nvfp4 requires Blackwell, so the run is requested on all three.

Note for reviewers

waives.txt carries 11 other waivers on this same test function under six unrelated bugs
(6384625, 5961814, 6422432, 6245394, 6474888, 6655987) covering different
moe_backend/parallelism/mtp_nextn combinations. Only the 6428063 line is touched here.

PR Checklist

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

[lori-ren]

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-1, GB300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-3, RTXPro6000D-PyTorch-Post-Merge-1"

[tensorrt-cicd]

PR_Github #68802 [ run ] triggered by Bot. Commit: 16e3762 Link to invocation

[tensorrt-cicd]

PR_Github #68802 [ run ] completed with state SUCCESS. Commit: 16e3762
/LLM/main/L0_MergeRequest_PR pipeline #56193 completed with status: 'UNSTABLE'

CI Report

⚠️ Multi-GPU Label Required:
Multi-GPU tests require the ci: full pre-merge approved label on this PR. Ask a member of NVIDIA/trt-llm-ci-approvers to add the label, then re-trigger CI with the same bot command (no rebase needed).

⚠️ Action Required:

• Please check the failed tests and fix your PR
• If you cannot view the failures, ask the CI triggerer to share details
• Once fixed, request an NVIDIA team member to trigger CI again

Link to invocation

[lori-ren]

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-1, GB300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-3, RTXPro6000D-PyTorch-Post-Merge-1"

[tensorrt-cicd]

PR_Github #69023 [ run ] triggered by Bot. Commit: 16e3762 Link to invocation

[tensorrt-cicd]

PR_Github #69023 [ run ] completed with state FAILURE. Commit: 16e3762
/LLM/main/L0_MergeRequest_PR pipeline #56397 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

• Please check the failed tests and fix your PR
• If you cannot view the failures, ask the CI triggerer to share details
• Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

[lori-ren]

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-1, GB300-4_GPUs-PyTorch-Post-Merge-2, GB300-4_GPUs-PyTorch-Post-Merge-3, RTXPro6000D-PyTorch-Post-Merge-1" --disable-fail-fast

[tensorrt-cicd]

PR_Github #69068 [ run ] triggered by Bot. Commit: 16e3762 Link to invocation

[tensorrt-cicd]

PR_Github #69068 [ run ] completed with state SUCCESS. Commit: 16e3762
/LLM/main/L0_MergeRequest_PR pipeline #56441 completed with status: 'SUCCESS'

CI Report

Link to invocation