Bump required R version to 4.0

[etiennebacher]

Necessary due to recent update of evaluate that has this requirement.

Close #500

[strengejacke]

Where do we need that package?

[etiennebacher]

Let's see if we follow this: easystats/easystats#401 (comment)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.20%. Comparing base (81dd0e0) to head (53952a0).
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #515      +/-   ##
==========================================
+ Coverage   91.14%   91.20%   +0.05%     
==========================================
  Files          76       76              
  Lines        6045     6086      +41     
==========================================
+ Hits         5510     5551      +41     
  Misses        535      535              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[strengejacke]

I think this is safe to do. This means that future updates from parameters will require minimum R4.0, too, when I update package dependencies. But: (foreign) packages that import either datawizard or parameters either also depend on R 4.0.0 or higher, or already import other packages that reqire R 4.0.0 or higher (e.g. package dotwhisker imports gtable, wich requires R 4.0). And furthermore, it's still possible to run all these packages in R 3.6 with older package versions. So, no serious concerns here, I think.

[etiennebacher]

Went back to the original discussion on the version policy, seems like the version policy is outdated because we can't test on 3.6 anymore?
https://easystats.github.io/easystats/articles/version_policy.html

[strengejacke]

I leave it to you to decide which R version should be required. It's always possible to run older package versions on R 3.6.

[etiennebacher]

Alright let's go with 4.0. We don't test on lower versions and it's available since May 2020.