Add a basic level of CI for this repo about nu_scripts

[amtoine]

@fdncred
if you have a look at the test stdlib command in the toolkit.nu of Nushell, you'll see a simple

cargo run -- -c $"use std testing; testing run-tests --path crates/nu-std ($extra_args)"

that's basically all there is to it

• you have some file somewhere
• you define commands in that file, possibly exported
• from the same file or another file, you add test commands with the #[test] annotation in their docstring
• you run

use std testing; testing run-tests --path ...

you can also filter the modules and the tests to run with the options of run-tests in case there are a bunch 👍

from where the tests are defined you just need to make sure the definitions you tests are in scope, especially when they come from another file 😋

[amtoine]

good idea 👍

two thinks that come to mind

• there might already be invalid scripts, i bet the CI won't be green directly
• even though we cannot test externals, we can check the syntax and parse everything to completion 😌

[fdncred]

Now that you mention it, it seems like it would be super valuable to have a script per nushell nu-command crate command. I'd guess that those scripts would probably be better in the nushell repo though and maybe tested through the std-lib test facility?

As we move toward 1.0, having every parameter on every command tested seems like a good idea. I'd like to start on this but I'm not sure how to use the std-lib testing facility. @amtoine is there a good testing model to look at somewhere for this?