Fix documentations to address CRAN NOTEs

[yoshidk6]

Address "Rd \link{} targets" issues raised in CRAN check results.
@noRd and @exportS3Method NULL were added for errors from roxygen2 regarding S3 methods (but better to consider exporting them?) r-lib/roxygen2#1589

There seem to be two other issues causing CMD Check errors:

• stan_betareg attr(X, "contrasts") not working properly (addressed in Use rstantools - Fix Build/Export Errors #625 ?)
• loo for poisson and negative binomial not meeting expect_equal(l$estimates, l2$estimates)

[jgabry]

Thanks for the PR! Looks OK at first glance, but I'll let @bgoodri also take a look.

[yoshidk6]

Thank you for the prompt response! May I also ask if the team is working toward addressing CRAN feedback (rstanarm to be archived by Sep 30)? I have a package that heavily relies on rstanarm (https://github.com/genentech/BayesERtools/) and would love to keep it on CRAN. I'm also happy to contribute if there is anything I can do to help.

[jgabry]

We won't let rstanarm get removed from CRAN. @bgoodri I assume you got an email from CRAN about this? When I look at the rstanarm results it seems like it's mainly some NOTES that need to be fixed.

[yoshidk6]

Fantastic to hear! I just run macbuilder check for this pull request version and and it seems to return clean.

• https://mac.r-project.org/macbuilder/results/1758217756-e2012ef1652217cb/

[jgabry]

Great! thanks for checking macbuilder.

[yoshidk6]

Hi! Wanted to check if there is any updates on rstanarm CRAN updates, as it's less than a week until the scheduled archival date? Would it help to open an issue?

[jgabry]

I've talked with @bgoodri (the maintainer) and he will submit it on time. Thanks for checking in! 

[yoshidk6]

Amazing thanks!!