[Docs] The big docs epic

[zth]

This issue is intended to track things we (desperately) need to add to the docs. Please feel free to comment on this issue, and I'll add to the list.

How the top level node field works.

• It always checks the cache (via id) first before going to the server. An example when this is useful is when you have a details page for a single instance of a type. Going via the cache first means Relay might be able to partially render a lot of the details view with the data that's currently in the cache, while fetching the rest of the data.
• If you use the node field to fetch only one type, RescriptRelay will "collapse" the response of node automatically, so you don't need to select the full union (select more than 1 type to opt out)
• Because of the benefits, it might be worth it considering using node for details pages (I always do that to leverage all of the benefits).

[redbar0n]

Resposting this tweet here:

I tried to find details of the benefits (the WHY) of #ReScriptRelay (esp. vs. normal #Relay) but I couldn't find any... not on the landing page, docs, or github page.. the closest I've found is some tweets of yours..

#rescript comp. is obv., but like to learn more!