no longer using dplyr_error

[romainfrancois]

dplyr 1.0.8 will stop using classed errors of type "dplyr_error".

This will also fail because of r-lib/testthat#1493, for example:

Failure (test-error-handling.R:11:3): `truth` should be factor
`sens(pathology, 1, factor("A"))` threw an error with unexpected message.
Expected match: "`truth` should be a factor"
Actual message: "Problem while computing `.estimate = metric_fn(truth = 1, estimate = factor(\"A\"), na_rm = na_rm, event_level = \"first\")`."
Backtrace:
  1. testthat::expect_error(sens(pathology, 1, factor("A")), "`truth` should be a factor")
       at test-error-handling.R:11:2
 17. yardstick metric_fn(truth = 1, estimate = factor("A"), na_rm = na_rm, event_level = "first")
 18. yardstick::metric_vec_template(...)
       at yardstick/R/class-sens.R:147:2
 19. yardstick::validate_truth_estimate_checks(...)
       at yardstick/R/template.R:174:2
 20. yardstick::validate_class(truth, "truth", cls[1])
       at yardstick/R/validation.R:201:2

Perhaps it's a good signal to start using snapshot tests here too.

[DavisVaughan]

I'll probably just make these expect_snapshot(expect_error())? I think that would avoid both issues right?

[romainfrancois]

Yeah I guess so.

[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.