add aggregation columns args

[tomjemmett]

• make sure uv is copying, not hardlinking
• removes unused build arg
• removes step thats not needed any more
• adds extra model argument: aggregation_columns

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (842132c) to head (ee56672).

Additional details and impacted files

@@            Coverage Diff            @@
##              main      #622   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           20        20           
  Lines         1131      1131           
  Branches        58        58           
=========================================
  Hits          1131      1131           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

✅ A new build is available.

You can use the following to use pull the image into your local environment:

docker pull ghcr.io/the-strategy-unit/nhp_model:pr-622

[Copilot]

Pull request overview

This PR extends the NHP demand model run pipeline to support configurable aggregation keys by introducing an aggregation_columns argument that influences how results are grouped, while also simplifying the Docker build and adjusting lint configuration to accommodate the updated function signatures.

Changes:

• Add aggregation_columns parameter plumbing through run_all / run_single_model_run and the concrete model constructors, and use it in Model.get_agg.
• Update unit tests to reflect the new argument being passed through, and adjust test fixtures for the new model state.
• Docker build tweaks (force uv to copy instead of hardlink; remove unused build arg / obsolete step) and loosen Ruff pylint max-args to 8.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 14 comments.

Show a summary per file

File	Description
src/nhp/model/model.py	Stores aggregation column configuration and uses it in get_agg groupby logic.
src/nhp/model/run.py	Threads aggregation_columns through run helpers; updates timeit to accept kwargs.
src/nhp/model/inpatients.py	Passes aggregation_columns into base Model initialisation.
src/nhp/model/outpatients.py	Passes aggregation_columns into base Model initialisation.
src/nhp/model/aae.py	Passes aggregation_columns into base Model initialisation.
tests/unit/nhp/model/test_run.py	Updates expectations for new argument propagation through run helpers.
tests/unit/nhp/model/test_model.py	Adjusts fixture patching to tolerate the updated Model.__init__ signature and aggregation state.
pyproject.toml	Increases Ruff pylint max-args to accommodate the new argument.
Dockerfile	Ensures uv links by copy, removes unused build arg and obsolete patch step.

[yiwen-h]

Thanks Tom! I'll review after copilot comments are addressed - ping me when done

[yiwen-h]

I can see how this would make things more flexible. Do we need to change code in nhp_model_databricks as well, as a result of these changes? Some concerns around whether or not this would break viewing of model results from ICB/national models in the outputs app

[tomjemmett]

We don't need to, but it would simplify things in that repo. Currently we have to insert some false values for some fields. This would allow us to not require them (e.g. sitetret)

[yiwen-h]

Thank you!