Weekly CI failure: H100 Mixtral 8x7B iteration-time mismatch

**Describe the bug**

The H100 weekly functional test `mixtral_8x7b_tp1pp4ep8vpp8_release` is intermittently failing in `ci-weekly` because the `iteration-time` metric is outside the approximate comparison tolerance.

Tag @NVIDIA/mcore-oncall to get oncall's attention to this issue.

**Failing run**

| Field | Value |
|-------|-------|
| Scheduled pipeline | `55286509` |
| Commit | `d1410e15` - Add MIMO runtime setup: per-role RNG seeding and DDP wrapping (#5285) |
| H100 child pipeline | `55287420` |
| Latest failing job | `344191894` |
| Job name | `mixtral_8x7b_tp1pp4ep8vpp8_release` |
| Test case | `tests/functional_tests/test_cases/mixtral/mixtral_8x7b_tp1pp4ep8vpp8_release` |

**History**

| Date | H100 child pipeline | Job ID | Status | Note |
|------|---------------------|--------|--------|------|
| 2026-05-26 | `52639682` | `326451132` | success | Last known pass before the intermittent failures |
| 2026-05-29 | `53136964` | `329741403` | failed | First recent failure |
| 2026-06-05 | `53873633` | `334879220` | failed | Same `iteration-time` mismatch |
| 2026-06-12 | `54641227` | `340436897` | success | Recovered temporarily |
| 2026-06-19 | `55287420` | `344191894` | failed | Failed again with the same signature |

**Error**

```text
tests/functional_tests/python_test_utils/test_pretraining_regular_pipeline.py::test_regular_pipeline

Actual values: 4.2905750000000005
Golden values: 3.8215149999999998
Approximate comparison of iteration-time: FAILED
APPROXIMATE test for metric lm loss: PASSED
APPROXIMATE test for metric mem-allocated-bytes: PASSED
APPROXIMATE test for metric mem-max-allocated-bytes: PASSED

AssertionError: The following metrics failed: iteration-time
```

The latest failing job also contains another extracted artifact with `actual 4.0858` versus `golden 3.821515`, so the symptom is variable H100 timing drift above the 5% `iteration-time` tolerance rather than a deterministic loss or memory mismatch.

**Steps/Code to reproduce bug**

Run the H100 release functional test for:

```bash
pytest tests/functional_tests/python_test_utils/test_pretraining_regular_pipeline.py::test_regular_pipeline
```

using:

```text
TRAINING_PARAMS_PATH=./tests/functional_tests/test_cases/mixtral/mixtral_8x7b_tp1pp4ep8vpp8_release/model_config.yaml
GOLDEN_VALUES_PATH=./tests/functional_tests/test_cases/mixtral/mixtral_8x7b_tp1pp4ep8vpp8_release/golden_values_dev_dgx_h100.json
```

**Additional context**

- This is separate from #5455, which tracks GPT3 weekly metric mismatches.
- Duplicate searches for `mixtral_8x7b_tp1pp4ep8vpp8 iteration-time` and `Mixtral H100 iteration-time approximate mismatch` did not find an open Mixtral-specific issue.
- CI system links are intentionally omitted; pipeline and job IDs are included for internal lookup.

Start with tests/functional_tests/python_test_utils/test_pretraining_regular_pipeline.py::test_regular_pipeline and the Mixtral model_config.yaml and golden_values_dev_dgx_h100.json paths. Run the supplied pytest command with those environment variables, then compare repeated H100 results to the 5% iteration-time tolerance while checking that loss and memory checks remain passing. Done means the intermittent mismatch is resolved and the release test passes reliably.

Written by the indexing model from the issue text.