Update augmentation pipeline

[melanibe]

This PR simplifies and generalizes the way we handle data augmentations:

1. The pipelining logic is now taken care of by a ImageTransformPipeline class that takes as input a list of transforms to chain together. This pipeline takes of applying transforms on 3D or 2D images. The user can choose to apply the same transformation for all channels or whether to apply different transformation for each channel (if each channel represents a different modality / time point for example).
2. The pipeline can now work directly with out of the box torchvision transform (as long as they support [..., C, H, W] inputs). This allows to get rid of nearly all of our custom augmentations functions.
3. The conversion from pipeline of image transformation to ScalarItemAugmentation is now taken care of under the hood, the user does not need to call this wrapper for each config class. This avoids having to call with ImagePipeline wrapper and then again another wrapper.
4. Getting rid of all the intrincated unused code for RandAugment & Co. The user has now instead complete freedom to specify the set of augmentations.