Agat

**Note: AGAT after this version (included) cannot load the well-trained model before.** If you need to do so, please use v8.0.5: https://pypi.org/project/agat/8.0.5/
- Fix bugs when traing model with voigt stress tensor.
- Add a node to edge layer: [agat/model/model.py](https://github.com/jzhang-github/AGAT/blob/v9.0.0/agat/model/model.py#L137-L138).
- Message passing: [agat/model/model.py](https://github.com/jzhang-github/AGAT/blob/v9.0.0/agat/model/model.py#L244-L247).
- Fix a bug when saving model [agat\lib\model_lib.py](https://github.com/jzhang-github/AGAT/blob/v9.0.0/agat/lib/model_lib.py#L161).
- Fix a bug when training model [agat\model\fit.py](https://github.com/jzhang-github/AGAT/blob/v9.0.0/agat/model/fit.py#L244-L252).

- Modify [agat/lib/model_lib.pyL40-L45](https://github.com/jzhang-github/AGAT/blob/v8.0.4/agat/lib/model_lib.py#L40-L45)
- Modify [agat/app/cata/high_throughput_predict.pyL208](https://github.com/jzhang-github/AGAT/blob/v8.0.4/agat/app/cata/high_throughput_predict.py#L208)

- Modify [agat/lib/model_lib.pyL178-L193](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/lib/model_lib.py#L178-L193)
- Add default parameter: `vasp_bash_path`[high_throughput_dft_calculation.pyL71](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/app/cata/high_throughput_dft_calculation.py#L71); [default_parameters.pyL242](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/default_parameters.py#L242).
- Modify `run_vasp()` function: [high_throughput_lib.pyL124-L149](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/lib/high_throughput_lib.py#L124-L149).
- Add transfer learning: [default_parameters.pyL97](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/default_parameters.py#L97). [agat/model/fit.pyL169-L174](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/model/fit.py#L169-L174)
- Add split graphs: [agat/data/build_dataset.pyL795-L824](https://github.com/jzhang-github/AGAT/blob/v8.0.3/agat/data/build_dataset.py#L795-L824)

- Convert TensorFlow to PyTorch backend.
- Updata docs.

- Add API for controling HP DFT calculation. [agat/default_parameters.py](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/default_parameters.py#L246-L251)
- Add `mask_reversed_magnetic_moments` in [agat/default_parameters.py](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/default_parameters.py#L58) and [agat/data/data.py](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py)
- Modify [agat/data/data.py](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py):
- Include stress in the graph: [agat/data/data.pyL273-L275](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py#L273-L275), [agat/data/data.pyL350-L352](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py#L350-L352).
- Update method of parsing the vasp data: [agat/data/data.pyL610](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py#L610), [agat/data/data.pyL625-L656](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py#L625-L656), [agat/data/data.pyL661-L675](https://github.com/jzhang-github/AGAT/tree/v7.14.0/agat/data/data.py#L661-L675).
- Update docs.

- Add [agat/app/cata/high_throughput_dft_calculation.py](https://github.com/jzhang-github/AGAT/tree/v7.13.4/agat/app/cata/high_throughput_dft_calculation.py).
- Shift atomic positions before fix bottom atoms: [agat/app/cata/high_throughput_predict.pyL225-L227](https://github.com/jzhang-github/AGAT/tree/v7.13.4/agat/app/cata/high_throughput_predict.py#L225-L227)
- Add `default_hp_dft_config` to [agat/default_parameters.pyL139-L246](https://github.com/jzhang-github/AGAT/tree/v7.13.4/agat/default_parameters.py#L139-L246).
- Add [agat/lib/HighThroughputLib.py](https://github.com/jzhang-github/AGAT/tree/v7.13.4/agat/lib/HighThroughputLib.py).
- Add [agat/lib/ModifyINCAR.py](https://github.com/jzhang-github/AGAT/tree/v7.13.4/agat/lib/ModifyINCAR.py).
- Upgrade docs.