add GLEM model, TAGDataset and example of GLEM (#9662)

reopened  #9591 

Feature summary:

- Add GLEM as GNN & LLM Co-training model to PyG
- adapt GLEM's LM to AutoModelForSequenceClassification from
transformers
- Lora support
- LM/LLM support
- ogbn-products/ogbn-arxiv testing finished
- TAGDataset can be used as a wrapper class for any node classification
dataset in PyG with LM tokenizer and associate raw text
- external prediction as pseudo labels supported

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Rishi Puri <puririshi98@berkeley.edu>
Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>

CHANGELOG.md

@@ -7,6 +7,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Added
 
+- Added `nn.models.GLEM` ([#9662](https://github.com/pyg-team/pytorch_geometric/pull/9662))
+- Added `TAGDataset` ([#9662](https://github.com/pyg-team/pytorch_geometric/pull/9662))
 - Added support for fast `Delaunay()` triangulation via the `torch_delaunay` package ([#9748](https://github.com/pyg-team/pytorch_geometric/pull/9748))
 - Added PyTorch 2.5 support ([#9779](https://github.com/pyg-team/pytorch_geometric/pull/9779), [#9779](https://github.com/pyg-team/pytorch_geometric/pull/9780))
 - Support 3D tetrahedral mesh elements of shape `[4, num_faces]` in the `FaceToEdge` transformation ([#9776](https://github.com/pyg-team/pytorch_geometric/pull/9776))

examples/llm/README.md

@@ -1,5 +1,6 @@
 # Examples for Co-training LLMs and GNNs
 
-| Example                              | Description                                                                                                                                                 |
-| ------------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| [`g_retriever.py`](./g_retriever.py) | Example for Retrieval-Augmented Generation (RAG) w/ GNN+LLM by co-training `LLAMA2` with `GAT` for answering questions based on knowledge graph information |
+| Example                              | Description                                                                                                                                                                                    |
+| ------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| [`g_retriever.py`](./g_retriever.py) | Example for Retrieval-Augmented Generation (RAG) w/ GNN+LLM by co-training `LLAMA2` with `GAT` for answering questions based on knowledge graph information                                    |
+| [`glem.py`](./glem.py)               | Example for [GLEM](https://arxiv.org/abs/2210.14709), a GNN+LLM co-training model via variational Expectation-Maximization (EM) framework on node classification tasks to achieve SOTA results |