docs(guides): add MLflow experiment tracking guide

[julpayne]

Summary

• Add a dedicated MLflow guide covering server config, experiment schema, per-benchmark runs, SDK backends (odh/upstream), CLEAR trace fetch vs result upload, and troubleshooting — verified against eval-hub, eval-hub-sdk, and eval-hub-contrib
• Cross-link related docs (home, quickstart, local mode, CLI, server API, system overview, MCP tools, CLEAR adapter) and fill gaps such as full mlflow.* config keys and CLEAR mlflow_traces_* parameters

Test plan

• Docs site builds successfully (npm run build)
• New sidebar entry Guides → MLflow opens /guides/mlflow/
• Links from home, quickstart, local mode, and CLEAR pages resolve
• Spot-check experiment tag examples match REST ({key,value}) vs MCP (map) behavior

Made with Cursor

Summary by CodeRabbit

• New Features

  • Added a comprehensive MLflow guide covering setup, experiment tracking, configuration, APIs, troubleshooting, and adapter integrations.
  • Added MLflow to the Guides navigation and homepage feature links.
  • Expanded IBM CLEAR documentation with MLflow trace input and result logging workflows.
  • Documented additional MLflow server settings, tags, run identifiers, and backend options.

• Documentation

  • Updated quick-start, CLI, local-mode, MCP, and architecture documentation with MLflow examples and guidance.

[coderabbitai]

Warning

Review limit reached

@julpayne, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 5 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 652b07d6-9a41-45f1-bd55-ac278d455449

📥 Commits

Reviewing files that changed from the base of the PR and between 4ddb5dc and 6d5626f.

📒 Files selected for processing (4)

• src/content/docs/adapters/clear/index.md
• src/content/docs/architecture/system-overview.md
• src/content/docs/getting-started/quickstart.mdx
• src/content/docs/guides/mlflow.mdx

📝 Walkthrough

Walkthrough

Adds a new MLflow guide and updates EvalHub documentation for tracking configuration, experiment submission, adapter callbacks, CLEAR trace ingestion, troubleshooting, navigation, and related references.

Changes

MLflow documentation

Layer / File(s)	Summary
MLflow tracking guide src/content/docs/guides/mlflow.mdx	Adds configuration guidance, experiment schemas, CLI/YAML/SDK/REST examples, logged metadata, adapter behavior, CLEAR usage, troubleshooting, and related links.
Adapter and CLEAR MLflow integration src/content/docs/adapters/clear/index.md, src/content/docs/guides/mlflow.mdx	Documents MLflow callback persistence, CLEAR trace retrieval, result uploads, filtering, and separate input/output experiment settings.
Configuration and submission references src/content/docs/architecture/system-overview.md, src/content/docs/reference/server-api.md, src/content/docs/getting-started/quickstart.mdx, src/content/docs/mcp/tools.md	Updates MLflow result fields, backend settings, server environment variables, experiment tags, and MCP/REST payload formats.
MLflow discovery and cross-links astro.config.mjs, src/content/docs/home.mdx, src/content/docs/guides/cli.mdx, src/content/docs/guides/local-mode.mdx, src/content/docs/guides/local-mode-tutorial.mdx	Adds navigation and homepage links and connects CLI and local-mode documentation to the MLflow guide.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• eval-hub/eval-hub.github.io#91: Updates overlapping MLflow documentation in the CLI and local-mode guides.

Suggested labels: kind/docs, documentation

Suggested reviewers: ruivieira, gnaulak-redhat

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description covers the summary and test plan but omits required template sections like What and why, Closes #, Type, and Breaking changes.	Reformat it to match the template: add What and why, Closes #, Type, Testing, and Breaking changes sections, even if some are briefly filled in.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely summarizes the main change: adding an MLflow experiment tracking guide.
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 unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch docs/mlflow-guide

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@src/content/docs/adapters/clear/index.md`:
- Line 65: Update the CLEAR adapter documentation to state that experiment.name
is the default experiment for logging results, while
parameters.mlflow_experiment_name overrides it when provided. Also revise the
workflow wording near the earlier experiment description to use this same
default-and-override behavior, while preserving the separate input-trace
setting.

In `@src/content/docs/getting-started/quickstart.mdx`:
- Around line 313-324: Update the Python quickstart example’s imports to include
JobSubmissionRequest, ModelConfig, and BenchmarkConfig alongside the existing
ExperimentConfig and ExperimentTag imports, so every model constructed in the
submission example is defined.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: e40905f7-a6a4-4c41-be0b-a25ec6462c6f

📥 Commits

Reviewing files that changed from the base of the PR and between 8de40b7 and 4ddb5dc.

📒 Files selected for processing (11)

• astro.config.mjs
• src/content/docs/adapters/clear/index.md
• src/content/docs/architecture/system-overview.md
• src/content/docs/getting-started/quickstart.mdx
• src/content/docs/guides/cli.mdx
• src/content/docs/guides/local-mode-tutorial.mdx
• src/content/docs/guides/local-mode.mdx
• src/content/docs/guides/mlflow.mdx
• src/content/docs/home.mdx
• src/content/docs/mcp/tools.md
• src/content/docs/reference/server-api.md