💡 Decide How We Plan to Calculate Node Features for All Dataloaders

[nithinmanoj10]

Not all of our dataloaders have the functionality to return the node features - which is very important for training a model on that dataset. So we should decide to make the following changes/additions

1. Add the function to return the node features of a dataset
2. Decide whether the node features that's returned must be normalized. Or should we return the data as it is and give the user the freedom to normalize the data however they want it. We should probably go with the latter.
3. Once these above changes are done, we should make sure to use the correct node features in the test script. Currently we are using randomly generates values as node features.

Issue #86 will also get addressed if we make the above changes