Adjustments to help bring package back to CRAN

[olivroy]

Fixes #117

tidyselect deprecation warnings.
I accepted snapshots with ggplot2 3.5.0
Updated gh actions with usethis::use_github_action().

There is one failure I can't seem to reproduce. You will see in the logs, but summary_slope seems to have lost one row, and I don't know if it is intentional o not.

I also removed unused packages from Suggests. (only leaving those used for examples, vignettes + tests) (readr is only used for creating a data in data-raw. markdown in not used, covr is used for the coverage (gh action installs it the same way the pkgdown gh action installs pkgdown)

Also removed what I assumed to be old vdiffr output.

Note that I see this failure locally, but I don't know what is causing it, nor if it is intended.

I don't understand the r-devel one, nor do I see it on Windows 4.3.2

[njtierney]

Wow, thank you @olivroy !

[olivroy]

@njtierney this should fix a first round of issues. (the cached Matrix version = 1.6-2) was the culprit for vignette build failures!

[njtierney]

Thanks so much again! I'm just going to run some checks locally, but I should be able to push this along very soon!

[olivroy]

Hi Nick, just wanted to remind you of this!

I have not had the chance to upgrade to R 4.4 yet, but if you have, maybe the failure is easier to reproduce locally?

[njtierney]

Thanks @olivroy ! This works locally for me, I'm not trusting GH actions right now, I've heard that there are a bunch of issues at the moment with GH actions.

Thanks again for all your help with this, and for the prompt! I had a small bug that I encountered, which should be solved now.

[njtierney]

Thanks again for your work on this @olivroy !

[njtierney]

Re

There is one failure I can't seem to reproduce. You will see in the logs, but summary_slope seems to have lost one row, and I don't know if it is intentional o not.

I believe this is just due to something random, this all looked fine to me otherwise.