refactor(dc_solver): start importer-backed preprocessing in powsybl example grids

[Copilot]

Checklist

Please check if the PR fulfills these requirements:

• PR Title follows conventional commit messages
• Tests for the changes have been added (for bug fixes / features)
• Docs have been added / updated (for bug fixes / features)
• All commits in this PR are DCO signed-off (see CONTRIBUTING.md)

Does this PR already have an issue describing the problem?

Tracked separately by the issue workflow.

What is the new behavior (if this is a feature change)?

Example-grid generation that previously relied on direct np.save mask writes now starts an importer-backed preprocessing path for powsybl examples, while preserving existing mask semantics for these synthetic grids.

• Powsybl preprocessing entrypoint

  • Added _run_importer_preprocessing_for_powsybl(...) in example_grids.py.
  • Uses UcteImporterParameters/CgmesImporterParameters (backend-appropriate shape) to invoke preprocessing.convert_file(...).
  • Handles unsupported synthetic-grid cases gracefully and falls back to explicit example mask writes.

• Mask persistence refactor

  • Replaced direct np.save(...) in:
    • case30_with_psts_powsybl
    • three_node_pst_example_folder_powsybl
  • Now uses save_numpy_filesystem(...) for mask persistence in these paths.

• Focused regression coverage

  • Added test_three_node_pst_example_folder_powsybl to assert relevant-node mask behavior remains aligned with PST-connected nodes.

_run_importer_preprocessing_for_powsybl(folder=folder, grid_file_path=grid_file_path, data_type="ucte")
save_numpy_filesystem(filesystem, masks_path / NETWORK_MASK_NAMES["relevant_subs"], rel_sub_mask)
save_numpy_filesystem(filesystem, masks_path / NETWORK_MASK_NAMES["trafo_pst_controllable"], trafo_mask)

Does this PR introduce a breaking change?

• Yes
• No

Original prompt

Start implementation

The user has attached the following file paths as relevant context:

• .github/copilot-instructions.md

[Chronological Review: The conversation began with the user requesting a refactor of functions that manually save data using `np.save`. The user specifically mentioned the function `parallel_pst_data_folder` as an example for this refactor. The user aimed to replace manual saving with a more structured approach using the `preprocessing.convert_file` function. Following this, the user requested a plan for refactoring all relevant functions. The most recent command was to start the implementation of the refactor.]

[Intent Mapping:

• User Request 1: "I want to refactor the example grid that still rely on manual use of np.save to save masks to disk."
• User Request 2: "Make sure to use proper parameters depending on whether it is a powsybl or a pandapower grid."
• User Request 3: "Start implementation."]

[Technical Inventory:

• Function parallel_pst_data_folder: A function that creates a preprocessed folder for parallel_pst_example(), runs the importer and preprocessing, and saves the network data.
• Function preprocessing.convert_file: A function intended to replace manual saving of data.
• Libraries: numpy, pypowsybl, pathlib, and custom modules related to toop_engine_dc_solver.]

[Code Archaeology:

• parallel_pst_data_folder: This function was discussed in detail, including its parameters and return type.
• Other functions mentioned: case30_with_psts_pandapower, case30_with_psts_powsybl, node_breaker_folder_powsybl, three_node_pst_example_folder_powsybl, and complex_grid_battery_hvdc_svc_3w_trafo_data_folder. Each of these functions involves similar data handling and saving processes.]

[Progress Assessment:

• Completed: The user has outlined the need for refactoring and provided an example function.
• Pending: The actual implementation of the refactor across all relevant functions.]

[Context Validation: All necessary context for continuing the refactor has been captured, including specific functions to modify and the intended use of preprocessing.convert_file.]

[Recent Commands Analysis:

• Last Agent Commands: The user executed a command to "Start implementation."
• Tool Results Summary: No specific tool results were returned as the command was to initiate implementation.
• Pre-Summary State: The agent was preparing to implement the refactor of the parallel_pst_data_folder function and related functions.
• Operation Context: The command was executed in the context of refactoring functions that currently use manual saving methods to utilize a more structured approach.]

1. Conversation Overview: - Primary Objectives: The user requested to refactor functions that rely on manual `np.save` for saving masks to disk, specifically mentioning the function `parallel_pst_data_folder` as an example. The user emphasized the need to use proper parameters for powsybl and pandapower grids. - Session Context: The conversation flowed from identifying the need for refactoring to outlining specific functions that require changes, culminating in a request to start the implementation. - User Intent Evolution: The user initially focused on identifying functions for refactoring and then moved to initiate the implementation phase.

2. Technical Foundation:

• parallel_pst_data_folder: A function that preprocesses data for parallel_pst_example(), runs the importer, and saves the network data.
• preprocessing.convert_file: A function intended to replace manual saving of data.
• Libraries: numpy, pypowsybl, pathlib, and custom modules related to toop_engine_dc_solver.

3. Codebase Status:

• parallel_pst_data_folder:
  • Purpose: Preprocesses data for parallel_pst_example().
  • Current State: The function is set to be refactored to use preprocessing.convert_file.
  • Key Code Segments: The function includes loading the network, running a DC load flow, and saving the grid file.
  • Dependencies: Relies on pypowsybl for load flow calculations and preprocessing for data conversion.

4. Problem Resolution:

• Issues Encountered: Manual saving methods in various functions.
• Solutions Implemented: The user has proposed a refactor to use preprocessing.convert_file.
• Debugging Context: No specific debugging issues were mentioned; the focus is on refactoring.
• Lessons Learned: The need for a structured approach to saving data has been identified.

5. Progress Tracking:

• Completed Tasks: The user has outlined the need for refactoring and provided an example function.
• Partially Complete Work: The implementation of the refactor is pending.
• Validated Outcomes: No specific outcomes have been validated yet as the implementation has just begun.

6. Active Work State:

• Current Focus: The user is focused on implementing the refactor of the parallel_pst_data_folder function and related functions.
• Recent Context: The l...

Created from VS Code.