Ensure mask matches resized image size

fix: correct invert_mask logic - RGB masks already have white=remove, only invert if invert_mask=False explicitly requested

fix: preserve image quality by limiting resize to 3000px, add output verification, improve debug logging

fix: handle RGB masks correctly - use RGB channels when alpha is all 255, so white pixels become white in mask

refactor: use exact reference model implementation from aryadytm/remove-photo-object - simple 255-alpha inversion

fix: improve image quality with LANCZOS4 interpolation, better mask validation, and high-quality PNG output

fix: match reference model exactly - use simple 255-alpha inversion like aryadytm/remove-photo-object

fix: add debug logs for mask processing; verify output changes to diagnose removal issues

fix: simplify mask processing to match reference model - direct white=remove detection

fix(mask): improve mask detection - handle RGB paint directly, detect magenta and white areas; match reference model behavior

fix(mask): ensure proper mask interpretation - selected areas are removed; add detailed logging

fix(mask): improve mask loading to match standard convention (white=remove); add debug logs

feat: invert-mask option in UI and core; add docs (no secrets)

Enable GPU auto-selection (CUDA if available) with DEVICE override; load TorchScript model on selected device. Reference: https://huggingface.co/spaces/LogicGoInfotechSpaces/Hair_stable_new/tree/main

Initial commit