Update README: multi-step is now the default

Default to 15-step multi-step sampling for sharp output; expose steps slider default 15

Delete files dinov2_encoder.py with huggingface_hub

Upload folder using huggingface_hub

Upload app.py with huggingface_hub

Upload folder using huggingface_hub

Upload dinov2_encoder.py with huggingface_hub

Upload models/dinov2_encoder.py with huggingface_hub

Upload folder using huggingface_hub

initial commit