Additional augmentations

Rather than reinventing the wheel one could just use `torchvision` transforms https://pytorch.org/docs/stable/torchvision/transforms.html

- [x] `Compose` (already recreated in `deepdow`)
- [ ] `RandomApply` - apply all with some probability
- [ ] `RandomChoice` - apply exactly one but at random
- [ ] `RandomOrder` - apply all but in random order
- [ ] `1Dwarping` - Affine would be a special case, one could in theory have any increasing function (derivative > 0)
- [ ] `RandomAffine` - scaling and translation along the y axis (lookback) could be a brilliant augmentation for `deepdow` tensors
- [ ] `RandomHorizontalFlip` - flipping the time flow, probably super confusing if one wants to pic up mean reversion
- [x] `Normalize` - a must together with some helper function that computes means, stds in the training set. However, it still assumes that the time series is stationary.
- [ ] `RandomErasing` - (similar to the current `Dropout` however it is contiguous regions)

Additionally, torchvision might be also helpful in other tasks (see #39)

The clear downside is introducing yet another dependency. Additionally, one might argue that it is better to go all the way and use `imgaug`, `albumentation`,...

Other nonvision augmentations:
- https://arxiv.org/abs/2002.12478
- https://github.com/arundo/tsaug
- https://halshs.archives-ouvertes.fr/halshs-01357973/document

Start with the torchvision transforms documentation linked in the issue and review the checked and unchecked augmentation list. The issue needs a decided scope and dependency direction before implementation; done would mean an agreed subset of augmentations is implemented for deepdow tensors with appropriate training-set helpers where required.

Written by the indexing model from the issue text.