Pre CRAN release 0.2.4

[clarkliming]

Summary

Prepares the 0.2.4 CRAN release and fixes the r-devel check failure reported on the CRAN check page.

The r-devel failure

tests/testthat/test-predict_counterfactual.R:73 errored on all five r-devel flavors (Debian clang/gcc, Fedora clang/gcc, Windows) with:

factor treatment has new levels trtpbo

while every r-patched / r-release flavor was OK, including Linux. So this is release vs. r-devel, not a platform difference.

R-devel changed model.frame.glm() to use the same logic as model.frame.lm() (PR#19036), which passes the fit's stored xlevels down to model.frame(). The test relabelled a treatment level after fitting and then handed the relabelled data back to predict_counterfactual(), so data no longer agreed with fit$xlevels. Release R let this through for glm fits only because the old model.frame.glm() re-evaluated glm(..., method = "model.frame") and dropped xlev along the way; lm fits have always errored on the same input.

The test now refits the model on the relabelled data so that the fit and the data agree, while still exercising a non-alphabetical treatment level order (trtpbo, trt1, trt2) — which is what the test is named for, and what the gl() fix from #111 guards.

Release prep

• Bump Version to 0.2.4 and Date to 2026-08-19.
• Rename the NEWS.md development heading to # RobinCar2 0.2.4, and add a ### Misc entry documenting the consequence for users: predict_counterfactual() now requires the treatment factor levels in data to match those of fit.
• Drop two unused inst/WORDLIST entries.

CRAN submission status

This release cannot be submitted yet — CRAN submissions are currently on hold. The CRAN submission portal states:

CRAN submissions will be offline from Aug 5, 2026 to Aug 19, 2026 -CRAN team vacation and maintenance work)

During this time, the submission of packages is not possible.

The stated window ends today, 2026-08-19, so the intention is to submit as soon as the portal reopens. In the meantime this PR and the v0.2.4-rc1 pre-release tag stage the submission candidate, so the tarball is ready to go without further changes once submissions resume.

Test plan

• devtools::test() on R 4.6.1 — all pass, 1 skip ({speff2trial} not installed locally).
• R CMD check --as-cran on R 4.6.1 (aarch64 macOS), with --no-vignettes --no-build-vignettes and _R_CHECK_FORCE_SUGGESTS_=false — checking tests ... OK and checking examples ... OK. The remaining ERROR/WARNINGs are all gaps in the local toolchain rather than package problems: no pandoc (so README.md/NEWS.md and the vignettes are unchecked), no pdflatex (so the PDF manual cannot be built), and an outdated HTML Tidy. CI covers those.
• lintr::lint_package() — no lints in the changed files.
• spelling::spell_check_package() — clean.

🤖 Generated with Claude Code

[github-actions]

Code Coverage Summary

Filename                     Stmts    Miss  Cover    Missing
-------------------------  -------  ------  -------  ------------
R/bias.R                        34       0  100.00%
R/find_data.R                    3       0  100.00%
R/predict_couterfactual.R       77       0  100.00%
R/prediction_cf.R               22       0  100.00%
R/robin_glm.R                   49       1  97.96%   40
R/robin_lm.R                    28       0  100.00%
R/surv_effect.R                 67       0  100.00%
R/survival_cov_adj.R           146       0  100.00%
R/survival_score.R             294       0  100.00%
R/survival.R                   345       1  99.71%   603
R/treatment_effect.R           101       1  99.01%   51
R/utils.R                      204       3  98.53%   83, 115, 119
R/variance_anhecova.R           44       0  100.00%
R/variance_hc.R                 10       0  100.00%
TOTAL                         1424       6  99.58%

Diff against main

Filename      Stmts    Miss  Cover
----------  -------  ------  --------
TOTAL             0       0  +100.00%

Results for commit: 6494915

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
survival	💚	$7.81$	$-1.39$	$0$	$0$	$0$	$0$

Results for commit 9d93d63

♻️ This comment has been updated with latest results.

[github-actions]

Unit Tests Summary

  1 files   13 suites   12s ⏱️
142 tests  96 ✅ 46 💤 0 ❌
348 runs  284 ✅ 64 💤 0 ❌

Results for commit 6494915.

♻️ This comment has been updated with latest results.

[danielinteractive]

@clarkliming I guess this now needs to be merged?