switch to {cli} in check_args() functions

[EmilHvitfeldt]

Ref: #1081

Now the call are being passed around, such as it is being emitted from fit() which feels appropriate. I wonder is there is a way to signal the model as well? so instead of saying Error in fit(): it says something like Error in fit() for discrim_linear(): or something to that liking. It would be nice for workflowsets

Also, I didn't detect any tests 😬 so I'll add those as well

library(parsnip)
library(discrim)

discrim_linear(penalty = -2) |>
  fit(Species ~ ., data = iris)
#> Error in `fit()`:
#> ! The amount of regularization, `penalty`, should be `>= 0`.

Extension packages:

• test {cli} changes to check_args() rules#83
• test {cli} changes to check_args() discrim#67
• test {cli} changes to check_args() baguette#75
• test {cli} changes to check_args() plsmod#38
• test {cli} changes to check_args() poissonreg#83
• test {cli} changes to check_args() censored#314

Empty tests

Some of the tests are left intentionally empty, this is either because there is a placeholder check_args() method with no checking, or no method at all. All missing methods will be handled by #1094

[EmilHvitfeldt]

This PR converts existing checks to {cli}, expanded checking will be tracked with #1095

[EmilHvitfeldt]

Todo: make sure we allow zero length input in input checking as per #1099

[simonpcouch]

I'm so happy you did this! These are a huge improvement.

I'm going to leave a "Comment" review to indicate that I'd like to take another look and re-review after you work through review comments, just because there are so many lines affected and cli markup mistakes can hard to spot visually on the first try. I'm around and interruptable today, so re-review should come quick. :)

[simonpcouch]

STELLAR

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.