Documentation: fill gaps in adapter docs, section on nodes vs. graphs, include available APIs in Key Features

[AdamGlustein]

During our csp hackathon we identified a few common issues in the documentation that many users faced.

• While there is documentation for writing adapters, there is no documentation for using adapters. We should prioritize completing that docs page as well as adding an example use of a csp input/output adapter inline with the docs (could be part of the Getting Started guide too).
• The principal point of confusion with new users is the difference between a node and a graph. I think we should devote an entire "Concepts" page to this difference to really solidify the idea, and also provide example incorrect code with explanations. For example, trying to call another node from a node is an extremely common error for beginners. Similarly is trying to perform a runtime computation in the graph.
• New users found it difficult to find function APIs in the documentation and often did not know the specific node they were trying to write already existed. It would be nice to include a mention under "Key Features" in the Wiki Home that there is a large set of pre-written nodes under baselib, math and stats. These should also be mentioned in the Getting Started guide once its ready.