Fix outstanding linting issues

[jamesmbaazam]

Summary

All open PRs were failing the lint-changed-files CI check because of pre-existing linting issues on main. Since that workflow only lints the files changed by each PR, the failures surfaced across different PRs depending on which files they touched. This PR resolves every genuine {lintr} issue in the package source and vignettes in one place so that, once merged, the other branches pass linting after rebasing onto main.

Issues addressed

Detected with lintr::lint_package() (using the repo's .lintr config with the etdev and cyclocomp packages):

• return_linter (28) — removed redundant explicit return() calls in R/borel.R, R/helpers.R, R/simulate.R, R/stat_likelihoods.R, R/utils.R and the vignettes, using implicit return instead.
• condition_call_linter (5) — added call. = FALSE to stop() calls in R/borel.R, R/helpers.R and R/stat_likelihoods.R.
• sprintf_linter (2) — replaced constant-string sprintf() calls with plain strings in R/epichains.R.
• object_overwrite_linter (1) — renamed the local dist variable (which shadowed the exported stats::dist) to simulated_stats in R/stat_likelihoods.R.
• sample_int_linter (4) — used sample.int(10, ...) instead of sample(1:10, ...) in vignettes/epichains.Rmd.
• one_call_pipe_linter (2) — removed single-call %>% pipes in vignettes/projecting_incidence.Rmd.

The remaining lints reported locally (object_usage_linter / missing_package_linter for epicontacts, truncdist, etc.) are environment-only false positives that clear on CI where those packages are installed.

Verification

• lintr::lint_package() reports 0 genuine lints after the changes.
• Full test suite passes (testthat::test_local()).

Checklist

• Added a NEWS.md item
• Changes are behaviour-preserving (explicit → implicit returns, internal rename, style-only fixes)
• Tests pass locally

[Copilot]

Pull request overview

This PR aims to eliminate pre-existing {lintr} findings on main so the lint-changed-files workflow stops failing unrelated open PRs, by applying behavior-preserving lint fixes across package R sources and vignettes.

Changes:

• Replaced redundant explicit return() calls with implicit returns in multiple R functions and vignette helper functions.
• Updated several stop() calls to include call. = FALSE, and adjusted vignette code to satisfy lintr rules (e.g., sample.int(), removing one-call pipes).
• Minor refactors to avoid shadowing (dist → simulated_stats) and remove unnecessary sprintf() usage in writeLines().

Reviewed changes

Copilot reviewed 10 out of 11 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
vignettes/projecting_incidence.Rmd	Removes redundant return() and replaces one-call %>% filter() with direct filter() calls.
vignettes/interventions.Rmd	Removes redundant return() from vignette helper functions.
vignettes/epichains.Rmd	Uses sample.int() and removes redundant return()s in vignette helper functions.
R/utils.R	Removes redundant explicit returns in internal helpers.
R/stat_likelihoods.R	Removes redundant explicit returns, adds call. = FALSE to stop(), and renames a local variable to avoid shadowing.
R/simulate.R	Removes redundant explicit returns in simulation functions.
R/helpers.R	Removes return(switch(...)) wrappers, adds call. = FALSE to stop(), and uses implicit returns.
R/epichains.R	Replaces constant-string sprintf() usage in writeLines() with plain strings.
R/borel.R	Removes redundant explicit returns and adds call. = FALSE to stop().
NEWS.md	Adds a NEWS entry noting linting issue resolution.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.