[None][test] Add DeepSeek V4 CI coverage

[lfr-0531]

Description

This PR updates DeepSeek V4 CI coverage and L0 multi-GPU triggering behavior:

• Adds GSM8K smoke evaluation to TestDeepSeekV4Flash.test_auto_dtype and TestDeepSeekV4FlashBase.test_auto_dtype, alongside the existing MMLU smoke checks.
• Disables pytest-threadleak for the sparse MLA forward test module because DSACacheManager creates background ThreadPoolExecutor threads, matching the existing DSA short-sequence test behavior.
• Makes L0 merge-request CI default to running multi-GPU tests without changed-file filtering, while preserving the explicit DISABLE_MULTI_GPU_TEST override.

Test Coverage

• pre-commit run --files jenkins/L0_MergeRequest.groovy tests/integration/defs/accuracy/test_llm_api_pytorch.py tests/unittest/_torch/attention/sparse/test_sparse_mla_forward.py
• .venv-3.12/bin/python -m py_compile tests/integration/defs/accuracy/test_llm_api_pytorch.py tests/unittest/_torch/attention/sparse/test_sparse_mla_forward.py
• Built and installed the PR wheel locally before the final test/CI edits. GPU pytest was not run after the user asked to avoid using shared GPUs.

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• 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.

[lfr-0531]

/bot run --add-multi-gpu-test --disable-fail-fast

[tensorrt-cicd]

PR_Github #47715 [ run ] triggered by Bot. Commit: 5a0bcac Link to invocation

[tensorrt-cicd]

PR_Github #47715 [ run ] completed with state SUCCESS. Commit: 5a0bcac
/LLM/main/L0_MergeRequest_PR pipeline #37614 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

Link to invocation

[lfr-0531]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #47743 [ run ] triggered by Bot. Commit: ea40cf4 Link to invocation

[tensorrt-cicd]

PR_Github #47743 [ run ] completed with state SUCCESS. Commit: ea40cf4
/LLM/main/L0_MergeRequest_PR pipeline #37638 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

Link to invocation

[lfr-0531]

/bot run

[tensorrt-cicd]

PR_Github #47814 [ run ] triggered by Bot. Commit: ea40cf4 Link to invocation

[tensorrt-cicd]

PR_Github #47814 [ run ] completed with state SUCCESS. Commit: ea40cf4
/LLM/main/L0_MergeRequest_PR pipeline #37706 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

Link to invocation

[lfr-0531]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #47927 [ run ] triggered by Bot. Commit: 9dcacb0 Link to invocation

[tensorrt-cicd]

PR_Github #47927 [ run ] completed with state SUCCESS. Commit: 9dcacb0
/LLM/main/L0_MergeRequest_PR pipeline #37774 completed with status: 'SUCCESS'

CI Report

Link to invocation