Optimized memory usage by reading only the required pixel window from the heightmap VRT instead of loading the entire dataset into memory. This prevents crashes on large datasets while maintaining scientific accuracy for ground-snapping.
