Accept posterior::draws objects in PPC/PPD functions

[ishaan-arora-1]

Closes #542

Summary

• validate_predictions() now checks for posterior::draws objects and coerces them to a plain matrix before the existing validation logic runs
• This mirrors what prepare_mcmc_array() already does on the MCMC side, so PPC/PPD functions now accept the same input formats as MCMC functions
• All existing ppc_* and ppd_* functions benefit since they all go through validate_predictions()

Changes

• R/helpers-ppc.R: added a posterior::is_draws() check at the top of validate_predictions(), coercing via as_draws_matrix() + unclass()
• tests/testthat/test-helpers-ppc.R: added tests for draws_matrix, draws_array, draws_df, and draws_rvars inputs, plus end-to-end tests with ppc_dens_overlay, ppc_hist, and ppc_stat

Test plan

• All existing PPC tests pass (510 pass, 0 fail)
• New tests cover all four draws formats
• End-to-end tests confirm plots are produced from draws input

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.15%. Comparing base (44c3e5c) to head (c3fc6f2).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #543   +/-   ##
=======================================
  Coverage   99.15%   99.15%           
=======================================
  Files          35       35           
  Lines        5920     5923    +3     
=======================================
+ Hits         5870     5873    +3     
  Misses         50       50           

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[jgabry]

Thanks! Yes we should allow posterior::draws objects. I'll try to review this later, I'm a bit short on time right this moment.

[jgabry]

I ended up having a moment to look at this now. Thanks for this. Just one change to make: I think the documentation for the yrep argument should be updated in man-roxygen/args-y-yrep.R.

I made some updates to the tests to make them more directly test what we care about. Otherwise this looks good aside from updating the doc.

[jgabry]

Looks good, thanks!