Autogl

- We have proposed NAS-Bench-Graph ([paper](https://openreview.net/pdf?id=bBff294gqLp), [code](https://github.com/THUMNLab/NAS-Bench-Graph), [tutorial](http://mn.cs.tsinghua.edu.cn/autogl/documentation/docfile/tutorial/t_nas_bench_graph.html)), the first NAS-benchmark for graphs published in NeurIPS'22. By using AutoGL together with NAS-Bench-Graph, the performance estimation process of GraphNAS algorithms can be greatly speeded up.
- We have supported the graph robustness algorithms in AutoGL, including graph structure engineering, robust GNNs and robust GraphNAS. See [robustness tutorial](http://mn.cs.tsinghua.edu.cn/autogl/documentation/docfile/tutorial/t_robust.html) for more details.
- We have supported graph self-supervised learning! See [self-supervised learning tutorial](http://mn.cs.tsinghua.edu.cn/autogl/documentation/docfile/tutorial/t_ssl_trainer.html) for more details.

- We have also released Chinese tutorial for the first time!
- AutoGL now support Deep Graph Library (DGL) backend to be interface-friendly for DGL users! All the homogeneous node classification task, link prediction task, and graph classification task are currently supported under DGL backend. AutoGL is also compatible with PyG 2.0 now.
- The heterogeneous node classification tasks are now supported!
- To make the library more flexible, the module model now supports decoupled to two additional sub-modules named encoder` and decoder
- AutoGL supports neural architecture search (NAS) to customize architectures for the given datasets and tasks.
- The link prediction task is now also supported!

Feel free to open issues and PR!

Dear all:

We are happy to release the toolkit we are developing: AutoGL, an autoML framework & toolkit for machine learning on graphs.

AutoGL is developed for researchers and developers to quickly conduct autoML on the graph datasets & tasks. Refer to our <a href="https://autogl.readthedocs.io">documentation</a> for more details. Please have fun playing with this first release and propose `issues` or `pull request`. Or contact us through <a href='mailto:autogl126.com'>autogl126.com</a>.

Known issues
1. OGB dataset is not fully supported, you may encounter bugs when conducting experiments on OGB datasets.
2. `AutoNE` in HPO now have some bugs in implementation.

Future work
1. Add NAS support
2. Add data/model training sampling support
3. Support more tasks, datasets and models