Fix bugs and improve pipeline flexibility

- Fix gpu_cleanup import missing in visualizations.py (NameError in workers)
- Fix t_pdf referenced before assignment when PDF is skipped
- Skip classification+DTM when DTM exists regardless of --force
- --force now only regenerates WebP/PDF, not classification/DTM
- --force-classification forces reclassification when needed
- Add laspy repair fallback for corrupt LAZ files (EVLR errors)
- Keep DTM TIF by default for reuse (--no-keep-tif to delete)
- Increase space between image and bottom cartouche (0.12→0.19)

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

lidar_pipeline/rendering.py

@@ -387,9 +387,9 @@ def tif_to_png(tif_file, vis_dir, resolution, keep_tif=False, source_info=None):
             # Fixed data area position — identical for ALL visualization types
             # This ensures overlay/superposition works across all WebP images
             data_left = 0.08
-            data_bottom = 0.12
+            data_bottom = 0.19
             data_width_frac = 0.74
-            data_height_frac = 0.78
+            data_height_frac = 0.71
 
             ax = fig.add_axes([data_left, data_bottom, data_width_frac, data_height_frac])
             if is_rgba or is_rgb: