Lama-cleaner

Add crop mode for very large image, this helps with memory issues.

<img width="846" alt="image" src="https://user-images.githubusercontent.com/3998421/159610609-0b935220-aae0-4fe2-8d84-fbe88a805514.png">

Add ldm model from: https://github.com/CompVis/latent-diffusion

|Original Image|LaMa|LDM|
|--------------|------|----|
||||
||||

- New UI
- Zoom & Pan

https://user-images.githubusercontent.com/3998421/153323093-b664bb68-2928-480b-b59b-7c1ee24a4507.mp4

Multi stroke support. Press and hold the cmd/ctrl key to enable multi stroke mode.

- To get high image quality, image is now resized by backend using opencv
- The resulting image is the same size as the original image
- Add resize radio button on web to select the size of the model forward. The output of the model will be different at different sizes, choose the size that suits you