Documentation for creating pipelines in Dagster

[MichaelHanksSF]

We need to have some documentation that gives a Python-familiar person:

• [ ] an understanding of the purpose of the pipelines, following a standard cleanfile, la-agg, pan-agg approach (nothing needed for more than this, like analysis code or sufficiency outputs etc.)
• [ ] an understanding of the structure of the code; what happens in what order; if a user was to create their own pipeline in this structure, what rough tasks would they do
• [ ] an introduction to any libraries that aren't likely to be familiar to a data science-y Python user; what the library does in the context of the code, where a user can find more information about it

[cyramic]

If we're going to document how to create pipelines, we'll also need to get into the following:

• [ ] How to make sensors to trigger jobs
• [ ] How to make assets to access locations