feat(ripley_template): Add Ripley‑L template, helpers, fixtures and t…

[fangliu117]

…ests
Ripley‑L template extracted from the NIDAP template, minus Foundry calls, A clean, importable template under spac.templates, Working unit tests:

1. template_utils.py – two tiny helpers:

• load_pickle → reads any pickle file (used for AnnData input).
• save_pickle → saves object and auto‑creates parent dirs.

2. ripley_l_template.py

• ._parse_params loads JSON from string, dict, or file.
• _validate enforces required keys and fills defaults.
• run_from_json loads AnnData → validates → calls spac.spatial_analysis.ripley_l with real parameters copied from the original template  → stores the DataFrame under adata.uns['ripley_l_results'] → pickles output.
• main block allows run the template from shell for ad‑hoc checks.

3. test_ripley_l_template.py –

• Creates a temp dir, writes the mock dataset, builds a params dict.
• Tests direct‑dict and JSON‑file execution.
• Asserts results are attached and output file exists.
• Cleans up automatically.

4. examples/ripley_l_params.json – ready‑made parameter file for manual runs or demos.

notes:

1. initial commit [eaf0e16]: CI is failing only because GitHub Actions can’t find the package under src/, so import spac breaks in the two new test‑files. Nothing is wrong with the code—just the workflow’s environment.
2. [3aa787d] tried to fix by installing the project editable inside the workflow, edit .github/workflows/gitflow-py-action.yml -> no help (The upstream action (fnlcr‑dmap/gitflow‑py/.github/workflows/parser.yml) starts its own container and
   overrides environment vars, so the PYTHONPATH set in the caller file never reaches the Python process that runs pytest.)
3. final commit [fb7343e] refactored the unit tests file by combining mock data and test into one -> unblock CI immediately, no need to fight the reusable workflow.

[Copilot]

Pull Request Overview

This PR introduces a new Ripley-L template for spatial analysis, providing a clean, platform-agnostic interface for running Ripley-L analysis with configurable parameters. The template is extracted from the NIDAP template and includes comprehensive testing infrastructure.

Key changes:

• New template system with parameter validation and flexible input handling (JSON file, string, or dict)
• Utility functions for pickle file operations with automatic directory creation
• Complete test suite with synthetic data fixtures for CI/CD integration

Reviewed Changes

Copilot reviewed 7 out of 8 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
src/spac/templates/ripley_l_template.py	Main template implementation with parameter parsing, validation, and Ripley-L analysis execution
src/spac/templates/template_utils.py	Utility functions for pickle file loading and saving operations
src/spac/templates/init.py	Package initialization with template documentation
tests/templates/test_ripley_l_template.py	Unit tests covering dict and JSON file parameter input scenarios
tests/templates/_fixtures.py	Mock data generation for testing with synthetic AnnData objects
examples/ripley_l_params.json	Example parameter configuration file for template usage
.github/workflows/gitflow-py-action.yml	CI configuration update to include src directory in PYTHONPATH

[Copilot]

[nitpick] The magic number 0.5 should be extracted as a named constant to improve code clarity and maintainability.

[Copilot]

[nitpick] The magic number 300 should be extracted as a named constant to improve code clarity and maintainability.

Suggested change

	adata.obsm["spatial"] = rng.random((n_cells, 2)) * 300
	adata.obsm["spatial"] = rng.random((n_cells, 2)) * SPATIAL_SCALING_FACTOR

[Copilot]

The file extension check using string operations is fragile. Consider using Path.suffix for more robust file extension detection.

Suggested change

	text = Path(src).read_text() if str(src).endswith(".json") else src
	text = Path(src).read_text() if Path(src).suffix.lower() == ".json" else src