Torch-points-kernels

Additions

- Cubic feature sampling kernel as proposed in https://arxiv.org/pdf/2006.03761

Bug fix
- Comilation with older GPU architecture

Additions
- Chamfer distance introduced in https://arxiv.org/pdf/1612.00603 for dense batches

Additions
- Windows support


Change
- Develop with python 3.7

Bug fix
- Fixed bug in region growing related to batching
- Ball query for partial dense data on GPU was returning only the first point. Fixed now

Additions
- Clustering algorithm for [PointGroup](https://arxiv.org/pdf/2004.01658.pdf)
- Instance IoU computation on CPU and GPU

Change
- Force no ninja for the compilation

Bug fix
- CPU version works for MacOS