feat: Implement landslide susceptibility mapping module

[vibhorjoshi]

• Add landslide susceptibility processing with tehsil-level clipping
• Implement MWS-level vectorization with 10 attributes per polygon
• Create Django REST API endpoint for on-demand generation
• Add GEE visualization script with interactive map and legend
• Implement comprehensive validation utilities (coverage, accuracy, attributes)
• Add unit tests with 6 test classes covering all components
• Provide 6 usage examples demonstrating all major features
• Update computing API and URLs for landslide endpoints
• Add path constant for pan-India landslide susceptibility asset

Features:

• 4-class susceptibility system (Low/Moderate/High/Very High)
• Polygon attributes: area by class, slope, curvature, LULC, score
• Async processing via Celery with GeoServer auto-publishing
• Follows existing CoRE Stack patterns (LULC, MWS architecture)
• Based on Mandal et al. (2024) methodology paper
• Production-ready with comprehensive documentation

Files added:

• computing/landslide/ module with 8 components
• docs/landslide_susceptibility.md (system documentation)
• LANDSLIDE_IMPLEMENTATION.md (implementation summary)
• LANDSLIDE_QUICK_REF.md (quick reference guide)
• IMPLEMENTATION_COMPLETE.md (achievement summary)

Files modified:

• computing/api.py (added generate_landslide_layer endpoint)
• computing/urls.py (added route)
• computing/path_constants.py (added constant)
• README.md (added table entry)

[vibhorjoshi]

@ankit-work7 please review this work

[vibhorjoshi]

@kapildadheech please review these changes and assign me this project

[Copilot]

Pull Request Overview

This PR implements a comprehensive landslide susceptibility mapping module for the CoRE Stack Backend, following the methodology from a research paper on geospatial modeling techniques for landslide susceptibility mapping.

Key Changes:

• Adds a new landslide susceptibility module with MWS-level vectorization and attribute computation
• Implements Django REST API endpoint for on-demand landslide susceptibility generation
• Provides extensive validation, visualization, and testing infrastructure

Reviewed Changes

Copilot reviewed 21 out of 21 changed files in this pull request and generated 15 comments.

Show a summary per file

File	Description
computing/landslide/landslide_vector.py	Core processing logic for clipping, vectorization, and export of landslide susceptibility data
computing/landslide/utils.py	Utility functions for statistics, validation, and visualization parameters
computing/landslide/validation.py	Quality assurance functions for coverage, attributes, and classification validation
computing/landslide/visualization.js	GEE Code Editor script for interactive visualization with legend
computing/landslide/tests.py	Unit tests covering vectorization, utils, validation, and API
computing/landslide/examples.py	Usage examples demonstrating API calls, validation, and statistics
computing/landslide/README.md	Module-level documentation with architecture, usage, and configuration
computing/landslide/__init__.py	Module initialization and overview documentation
computing/api.py	New REST API endpoint generate_landslide_layer for async task triggering
computing/urls.py	URL routing for the new landslide API endpoint
computing/path_constants.py	Asset path constant for pan-India landslide susceptibility
docs/landslide_susceptibility.md	System-level documentation with methodology, use cases, and performance guidelines
README.md	Updated main README with landslide module entry in script path table
gee_kyl/process_landslide_susceptibility.py	Standalone GEE processing script for landslide susceptibility computation
gee_kyl/tests/test_process_import.py	Basic test for the standalone script
gee_kyl/ee_code_editor.js	Simple visualization helper for GEE Code Editor
gee_kyl/requirements.txt	Python dependencies for the standalone script
LANDSLIDE_IMPLEMENTATION.md	Detailed implementation summary with acceptance criteria review
LANDSLIDE_QUICK_REF.md	Quick reference guide for common tasks
IMPLEMENTATION_COMPLETE.md	Executive summary of the implementation

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The loop iterates through SUSCEPTIBILITY_CLASSES and calls reduceRegions() for each class, but each iteration overwrites the previous "sum" property. This will cause all intermediate area calculations to be lost, with only the last class area being preserved under the key "sum".

Instead, the results of reduceRegions() should be stored with unique property names for each class within the loop, or the property should be renamed immediately after each reduceRegions() call before the next iteration. Currently, the add_area_ha function tries to access feature.get("sum"), but "sum" gets overwritten in each iteration.

Consider using a pattern where the area for each class is computed and renamed to the final property name (e.g., {class_name}_area_ha) before moving to the next iteration.

Suggested change

	# Convert from m² to hectares and add as property
	def add_area_ha(feature):
	area_m2 = feature.get("sum")
	area_ha = ee.Number(area_m2).divide(10000)
	return feature.set(f"{class_name}_area_ha", area_ha)
	fc = fc.map(add_area_ha)
	# Convert from m² to hectares and add as property, then remove "sum"
	def add_area_ha(feature):
	area_m2 = feature.get("sum")
	area_ha = ee.Number(area_m2).divide(10000)
	# Remove "sum" property after storing area
	return feature.set(f"{class_name}_area_ha", area_ha).remove("sum")
	fc = fc.map(add_area_ha)

[Copilot]

[nitpick] Unnecessary blank lines at the beginning of the file. Consider removing lines 1-4 to follow standard Python style conventions.

Suggested change

[vibhorjoshi]

@copilot open a new pull request to apply changes based on the comments in this thread

[aaditeshwar]

Hi all, Do please join the discord channel and googlegroup linked from this announcement of the CoRE stack innovation challenge, and do participate in the challenge too: https://core-stack.org/core-stack-innovation-challenge-1st-edition/. Our first community call on Friday 3-4pm will be very useful to get started.