Revised documentation system

[topepo]

This PR revamps the documentation in a few ways:

• The engine list on help packages is not dynamic; it lists all known engines.
• The list shows when an extension package might be needed.
• The engine documentation has information on which mode needs an extension package.
• parsnip now has a file of all known models/modes/engines for models that are defined in parsnip.
• Numerical superscripts are used to make notes in the engine list.
• Contributors should not have to modify the underlying engine docs for most changes.

See inst/README-DOCS.md for an overview of the documentation process. To test it out:

1. Install the devel versions of the packages listed by parsnip:::extensions().
2. Run knit_engine_docs()
3. Build the Rd files via devtools::document()
4. Reinstall parsnip (I would avoid load_all() here)
5. Take a look at the man pages.

I've run it through pkgdown and the superscripts don't cause any issues.

Open question: should we commit the md files too?

[juliasilge]

Here is what is still "to do" from my perspective:

• This PR adds a file that I can't tell what it is for: example_mlm.Rmd. Can we remove it from this PR?
• Make a template for the linear predictor values (it's repeated)
• There is documentation in the file engine_docs.R that I believe is outdated; check out this comment and others.
• It looks to me like we can avoid the "May require a parsnip extension package." message. I think this is a high priority because it is confusing.
• The way this sentence gets put together is confusing and I suggested some alternatives.
• I find this pretty confusing (used in several files):

gee() always prints out warnings and output even when silent = TRUE. When using the gee engine, it will never produce output, even if silent = FALSE.

How about something like this:

The gee::gee() function always prints out warnings and output even when silent = TRUE. The parsnip "gee" engine, by contrast, silences all console output coming from gee::gee(), even if silent = FALSE.

Is that accurate?

I still haven't successfully run knit_engine_docs() but I'll keep trying. I did eventually get this to run, and then I could run devtools::document().

[juliasilge]

Basically me about those ⬆️ issues:

[juliasilge]

Can we in this PR add a release question to remind us to knit_engine_docs()?

[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.