Graphchem

- updates PyTorch to 2.4.0
- adds MANIFEST.in for including CSV datasets with GraphChem
- bump version to 2.1.3

- `graphchem.preprocessing.MoleculeEncoder` how has the `save` and `load` methods
- `graphchem.preprocessing.load_encoder` allows users to load pre-saved MoleculeEncoder objects
- Build/installation now relies on `pyproject.toml` instead of the deprecated `setup.py`
- Added a GitHub action workflow to publish GraphChem to PyPI whenever a new version is published

Use pre-built RDKit binaries
Added readthedocs API documentation
Preparing for Azure unit testing

- New method for atom/bond tokenizing (integer tokens, passed through torch.nn.Embedding layer at start of model)
- Removed GRU and scatter add functions, leads to faster and more accurate results
- Changed MFConv to GeneralConv (GeneralConv considers both atom and bond features during its message passing step)
- General flow for each message passing step is:
- embedded_bonds -> EdgeConv -> updated embedded_bonds
- embedded_atoms + embedded_bonds -> GeneralConv -> updated embedded_atoms
- Separate edge/node convolutions, where node convolutions also account for updated edge states, is used effectively by Jørgensen et al. ([paper](https://doi.org/10.48550/arXiv.1806.03146))
- Updated/added examples

This update overhauls GraphChem. The goal was to make GraphChem more "Torchy" in nature - for example, removing the CompoundOperator and giving control of training back to the user.

While this reduces the base functionality of GraphChem, ultimately its new construction is more modular and will play nicer with PyTorch.

- GCN now uses a GRU layer for edge network, in addition to existing GRU for node network
- Connectivity algorithm fixes, considers reverse case and updates connectivity indices accordingly
- Included example scripts/notebooks for CN, RON, and MON