feat(experiments): Expose end-to-end latency metric on Evaluations

[shanaiabuggy]

Summary

Adds a first-class end-to-end latency metric to Evaluations: the total time to run an Evaluation's tasks serially. It is the sum of per-test-case latency, where a test case run more than once contributes the average of its attempts. Switchyard requested this for the Experiments leaderboard.

Under the hood this is latency_ms.sum — the rollup already computes latency test-case-weighted (per-attempt → averaged per test case → distribution across test cases), so latency_ms.sum already carries exactly this meaning. This PR names it as a dedicated response field and surfaces it in Studio; no new aggregation was required.

Related Issue

Relates to ASE-805 (Linear).

Changes

• Intake API: add an end_to_end_latency_ms computed field to EvaluationResponse (= latency_ms.sum), typed nullable (null when no session carries latency).
• Contract: regenerated OpenAPI (platform specs) and SDKs — Python Optional[float], web number | null.
• Studio: new sortable + filterable "End-to-end latency" column in the Experiments leaderboard (ExperimentDataView), mapped to the API's latency_ms.sum for sort/filter, plus the reverse default_sort mapping so a saved -latency_ms.sum highlights this column. Null-safe cell.
• Tests: test_end_to_end_latency.py — the field equals latency_ms.sum and serializes; null when no latency.

Type of Change

• Code change (feature, bug fix, or refactor)
• Code change with documentation updates
• Documentation only
• Contributor tooling or automation
• CI, build, or test infrastructure

Quality Gates

• Tests added or updated for changed behavior
• Existing tests cover changed behavior — justification:
• Tests not applicable — justification:
• Documentation not applicable — justification: the response field carries an inline OpenAPI description; the end-user Intake/Experiments docs live outside this repo.

Verification

• Pull request title follows the repository's Conventional Commit format
• Every commit includes an appropriate Signed-off-by: trailer
• uv run pre-commit run -a passes, or any blocked checks are identified below
• Targeted tests pass, or tests are marked not applicable above
• No secrets, API keys, or credentials are included

Targeted validation:

• uv run --frozen pytest services/intake/tests/test_end_to_end_latency.py -q → 2 passed
• pnpm --filter nemo-studio-ui typecheck → passes
• pnpm --filter nemo-studio-ui lint → clean
• pnpm --filter nemo-studio-ui test src/components/dataViews/ExperimentDataView/util.test.ts → 5 passed

🤖 Generated with Claude Code

Summary by CodeRabbit

• New Features
  • Added an end-to-end latency column to experiment data views.
  • Enabled sorting and numeric filtering by latency.
  • Included latency in saved and default view configurations.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 02e51d22-3b64-4170-a3eb-954d29f01849

📥 Commits

Reviewing files that changed from the base of the PR and between 63142e4 and ba60e4c.

📒 Files selected for processing (1)

• web/packages/studio/src/components/dataViews/ExperimentDataView/index.tsx

🚧 Files skipped from review as they are similar to previous changes (1)

• web/packages/studio/src/components/dataViews/ExperimentDataView/index.tsx

---

📝 Walkthrough

Walkthrough

Changes

Evaluation latency

Layer / File(s)	Summary
Studio latency integration web/packages/studio/src/components/dataViews/ExperimentDataView/index.tsx	Adds an end-to-end latency column with duration formatting, sorting, numeric filtering, and mappings to the latency_ms.sum API field.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes exposing the end-to-end latency metric on Evaluations, which is the main change.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch sbuggy/ase-805-expose-end-to-end-serial-latency-metric-on-evaluations

---

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

[github-actions]

Suite	Lines Covered	Line Rate	Branch Rate
Unit Tests	31436/40061	78.5%	63.0%
Integration Tests	18303/38013	48.1%	20.8%