"Contributing to Rover", aggregate issue

[lrlna]

This issue is encapsulate a few items that we talked about in the past few weeks regarding making Rover more friendly to contribute to.

Documentation items:

1. Create a CONTRIBUTING.md that includes

   • building Rover from source on various environments
   • this should point people to read ARCHITECTURE.md
   • local workspace good-to-haves (vscode, rust-analyzer, cargo fmt, clippy) and how to set those up to make sure that code written locally will pass CI on remote (for example, not having cargo fmt run on-save in vscode may cause problems)
   • what it looks like to add a command from scratch
     • adding a command in to Rover's src
     • adding an endpoint in rover_client
     • setting up a separate RoverClientError (and when it's necessary to do so)
     • adding a Suggestion and Code to errors (including creating an error file)
     • adding a different output type to RoverStdout

2. Create ARCHITECTURE.md that includes

   • how our crates work together
   • how we build for alpine (no composition)
   • how our error setup works
   • our thought process on command/flag creation language and how we expect future commands to follow that
   • documentation on rover client
   • how our build process works
   • how our testing works
   • how our build.rs works

3. cargo xtask Use cargo-xtask for automating builds #388

4. any sort of cleanup we would like to do in rover_client RFC: Improve architecture of rover-client #142

5. Adding rust docs to any public facing APIs

6. Adding the following to public facing crates:

   #![forbid(unsafe_code)]
   #![deny(missing_debug_implementations, nonstandard_style)]
   #![warn(missing_docs, future_incompatible, unreachable_pub, rust_2018_idioms)]

[ndintenfass]

Probably worth adding to the list is some way to talk about how to do good command design and how we decide whether something should be a top level command vs. a sub-command (along with style guide on when to use a positional vs flagged input).

[EverlastingBugstopper]

@lrlna do you think we can close this and maybe open a new issue for docs stuff? We've done everything in this issue except 4, 5, and 6

[lrlna]

yea, let's close this!