Spatial stats

[AnthonyChristidis]

Spatial Statistics Integration (Squidpy / PySAL)

This PR completely integrates spatial statistics into the oyLabImaging pipeline, allowing users to move beyond temporal single-cell tracking to analyze spatiotemporal cell-cell interactions.

Key Features Added to Results.py:

• Global Spatial Autocorrelation: Added Moran's I and Geary's C to measure global clustering of fluorescent markers across the dish.
• Local Spatial Hotspots: Added Local Moran's I and Local Bivariate Moran's I. Calculates a per-cell spatial score to identify High-High (Hotspots), Low-Low (Coldspots), and boundary cells. These scores are automatically saved to P.framelabels[t].regionprops.
• Neighborhood Enrichment: Categorical neighborhood analysis (sq.gr.nhood_enrichment). Users can define a cluster_key to see which discrete cell states (e.g., Infected vs Healthy) physically touch more than random chance.
• Dynamic Plotting: Added R.plot_spatial_stats() to automatically generate X/Y spatial maps, grouped bar charts (for single timepoints), and continuous line graphs (for timelapse).
• Interactive Napari Integration: Added R.show_spatial_map_napari() to seamlessly overlay the calculated spatial hotspots directly onto the raw microscopy TIFFs. Includes support for full-movie timelapse sliders and stacked toggleable layers.

Dependencies Added (setup.py & environment.yml):

• squidpy
• anndata
• esda
• libpysal

Documentation:
Added two fully documented Jupyter Notebooks to the notebooks/ directory demonstrating the complete pipeline on real data:

1. SpatialStats_SingleTimepoint.ipynb
2. SpatialStats_Timelapse.ipynb