feat(selection): separate selection geometry from image content processing

- Introduced a new design for selection rasterization that decouples the selection geometry from the image content, ensuring that the SelectionMask is independent of the source image's alpha and color.
- Updated the `rasterize_selection_path` function to only accept selection paths and image dimensions, removing parameters related to source alpha and alpha thresholds.
- Implemented a new approach for handling selection paths using non-zero winding fill, ensuring overlapping areas are preserved.
- Enhanced performance by utilizing a scanline algorithm for rasterization, significantly reducing processing time for high-resolution images.
- Updated internal documentation and added comprehensive tests to validate the new behavior and performance improvements.
- Ensured backward compatibility by maintaining existing interfaces where necessary while introducing breaking changes in the selection processing logic.

src/anyimage/server/jobs/cutout.py

@@ -1,4 +1,4 @@
-"""Generate optional AI files for one Blender-rasterized Cutout Selection."""
+"""Generate optional AI files from one rectangular Cutout image input."""
 
 import json
 from pathlib import Path

src/anyimage/server/jobs/depth_plane.py

@@ -16,6 +16,7 @@ def run(context, parameters):
             parameters,
             images[0],
             depth_mode="Z",
+            normal_mode="TANGENT",
         )
     context.check_cancelled()
     context.progress(1.0, "Depth Plane data is ready")

src/anyimage/server/selection/refine_selection.py

@@ -5,14 +5,16 @@ def _crop_to_alpha(image):
     import numpy as np
 
     pixels = np.asarray(image.convert("RGBA"), dtype=np.uint8)
-    visible_y, visible_x = np.nonzero(pixels[:, :, 3] > 0)
+    visible = pixels[:, :, 3] > 0
+    visible_y = np.flatnonzero(np.any(visible, axis=1))
+    visible_x = np.flatnonzero(np.any(visible, axis=0))
     if not len(visible_x):
         raise ValueError("The generated Selection contains no visible pixels")
     bounds = (
-        int(visible_x.min()),
-        int(visible_y.min()),
-        int(visible_x.max()) + 1,
-        int(visible_y.max()) + 1,
+        int(visible_x[0]),
+        int(visible_y[0]),
+        int(visible_x[-1]) + 1,
+        int(visible_y[-1]) + 1,
     )
     return image.crop(bounds), bounds
 
@@ -24,7 +26,7 @@ def refine_selection(
     device,
     output_path,
 ):
-    """Refine an image Selection from an already rasterized Selection."""
+    """Refine foreground Alpha within one rectangular Selection context."""
     model_manager = context.resource("model_manager")
     foreground, _load_ms, _inference_ms = infer_foreground(
         input_path,