[Feature request] better overview and examples of smaller scale models

[colin2328]

request on behalf of @adamlerer (adding here to be able to reference)

1. the readme contains no information about how to actually use torchrec, or a problem description of what problem it's trying to solve. It instead gives a list of advanced engineering features that aren't related to the high-level task https://github.com/pytorch/torchrec
2. Next I click to docs, which gives an API description but still no explanation of what problem it's trying to solve or how to actually use it
3. Next I get to the tutorial, which kind of tells me how to use it. But do I really need to set up distributed torch and stuff for that?

Here's a simple setup to train a recommender system that you can get running in an hour and will give you good results. And don't worry, we've built lots of fancy features so that it will scale up to huge data sizes