Napari-n2v

- Relax constraints on dependencies https://github.com/juglab/napari-n2v/pull/52

- Improve bioimage.io export/import and validation
- Add export/import of TF SavedModel format
- Add possibility to train without augmentation https://github.com/juglab/napari-n2v/issues/32
- Fix missing icons https://github.com/juglab/napari-n2v/issues/33
- Allow more validation patches https://github.com/juglab/napari-n2v/issues/34
- Fix bug relative to axes widget update https://github.com/juglab/napari-n2v/issues/35
- Fix prediction scale https://github.com/juglab/napari-n2v/issues/36

Update setup

Fix Demo plugin.

Update Readme for PyPi and add citations.

First napari-n2v release!

The plugin contains:
- Loading of data from napari layers or disk
- Training and re-training
- Expert settings for advanced users
- TensorBoard to follow training
- Prediction from napari layers or disk
- Export of models to TensorFlow or bioimage.io formats
- Documentation
- Tutorials