Pas d'interpolation dans le DTM: les zones sans données restent NaN

- Suppression de l'interpolation NearestNDInterpolator dans create_dtm_fast
- Les pixels sans données LiDAR restent NaN dans le DTM et les
  visualisations — pas de valeurs fictives qui faussent les calculs
- nodata=float('nan') dans le GeoTIFF de sortie pour identifier les vides
- _save_tif() détecte automatiquement les NaN et écrit le flag nodata

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

lidar_pipeline/visualizations.py

@@ -25,14 +25,19 @@ else:
     xp = np
 
 
-def _save_tif(output_path, data, transform, crs, dtype='float32', count=1):
+def _save_tif(output_path, data, transform, crs, dtype='float32', count=1, nodata=None):
     """Helper to save a 2D or 3D array as GeoTIFF."""
+    # Auto-detect nodata for float types with NaN
+    if nodata is None and dtype.startswith('float') and np.any(np.isnan(data)):
+        nodata = float('nan')
+
     if data.ndim == 2:
         height, width = data.shape
         with rasterio.open(
             output_path, 'w', driver='GTiff',
             height=height, width=width, count=count,
-            dtype=dtype, crs=crs, transform=transform, compress='lzw'
+            dtype=dtype, crs=crs, transform=transform,
+            compress='lzw', nodata=nodata
         ) as dst:
             dst.write(data.astype(dtype), 1)
     elif data.ndim == 3:
@@ -40,7 +45,8 @@ def _save_tif(output_path, data, transform, crs, dtype='float32', count=1):
         with rasterio.open(
             output_path, 'w', driver='GTiff',
             height=height, width=width, count=bands,
-            dtype=dtype, crs=crs, transform=transform, compress='lzw'
+            dtype=dtype, crs=crs, transform=transform,
+            compress='lzw', nodata=nodata
         ) as dst:
             for i in range(bands):
                 dst.write(data[i].astype(dtype), i + 1)