feat: update examples/pipeline_2 to align with package better

[pikes-ons]

Updated the example pipeline 2 code to account for correct logging and run_directory location searches. Also provides stage_results. This example should now be an idea example for how to run a Pipeline in this package with full utilisation.

[Copilot]

🟡 Changes recommended

The updated example stages now require an ExecutionContext but still allow/trigger context=None execution paths that will fail at runtime (including an explicit main(context=None) in 0_clean_data.py).

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

This PR updates the examples/pipeline_2 example pipeline to rely on the package’s ExecutionContext for resolving input/output locations (run-directory based), and to return richer per-stage outputs (“stage_results”) rather than using fixed file paths.

Changes:

• Switch stage scripts to compute IO paths via context (e.g., resolve_output_root(), resolve_given_path()), aligning outputs with the pipeline run directory.
• Add structured return dictionaries from stages (e.g., record counts, output locations, summary metrics).
• Update conf.yaml to use output_dir: examples/pipeline_2/runs and remove hardcoded stage IO locations; remove pre-generated intermediate CSVs from data/.

File summaries

File	Description
examples/pipeline_2/scripts/0_clean_data.py	Writes cleaned data into the run directory and returns stage outputs.
examples/pipeline_2/scripts/1_derive_vars.py	Reads prior-stage output via context resolution and writes derived dataset to the run directory with outputs returned.
examples/pipeline_2/scripts/2_reporting.py	Resolves prior-stage dataset via context and writes report into the run directory; returns summary outputs.
examples/pipeline_2/data/orders_prepped.csv	Removes committed intermediate output artifact.
examples/pipeline_2/data/orders_cleaned.csv	Removes committed intermediate output artifact.
examples/pipeline_2/conf.yaml	Updates output_dir to runs and removes per-stage input/output location entries.

Review details

Suppressed comments (1)

examples/pipeline_2/scripts/0_clean_data.py:69

• The __main__ entrypoint explicitly calls main(context=None), but this stage now requires an ExecutionContext; running the file directly will always fail. Prefer failing fast with a clear message (or wiring up a real context).

if __name__ == "__main__":
    main(context=None)

• Files reviewed: 6/6 changed files
• Comments generated: 3
• Review effort level: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

🟡 Changes recommended

The updated example stage entrypoints now require an ExecutionContext but still have direct-run invocation paths that will fail (TypeError/AttributeError) without a clearer handling of missing context.

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Review details

• Files reviewed: 6/6 changed files
• Comments generated: 3
• Review effort level: Lite