Add unified results for completed Slurm scenarios

[podkidyshev]

Ordinary Slurm scenarios write a single experiment.json in the scenario results directory when execution finishes or fails. The file follows API Schema v0.2's full Experiment shape and contains test cases, actual submitted runs, scheduler/workload status, timing, and canonical metrics.

The runner captures results before advancing mutable test iterations. NCCL and NIXLBench use their existing metric_observations() hooks; test-level measurements average successful iterations at matching dimension points, while each run keeps its original measurements. Output is replaced atomically, and extraction or output failures warn without changing execution behavior. Timestamps without timezone information remain null.

This first deliverable covers completed normal Slurm execution. Live updates, DSE, single-sbatch, additional workload metric extractors, and programmable API changes are deferred.

Validation:

• 225 affected tests passed, including three new focused test functions (five cases).
• Generated experiment files validated against the live API Schema v0.2, including date-time formats.
• Full Pyright, Ruff, import contracts, dead-code checks, and commit hooks passed.
• Slurm execution is covered with mocks; no cluster run was performed.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

---

_{Comment @coderabbitai help to get the list of available commands.}