docs: fix text/documentation drift and broken examples (no behavior change)

[vertesy]

Motivation

PR 1 of a batched repo audit (per the attached Repository Update Workflow Instructions). Pure documentation/text fixes — easiest category to review, zero behavior change. Full audit categorized ~40 problems; this PR covers the "text/documentation" bucket. Follow-up PRs (2-7) will cover undeclared-dependency crashes, a 1:length(x) style/edge-case sweep, and other logical bugs, one focused PR at a time.

What's in this PR

• README: fixed a doc-generator bug — item #115 was mislabeled which.max.multi() (a private nested helper, not public API) while carrying get_max_colname_per_row()'s real description, which was missing from the list entirely.
• DESCRIPTION / Development/config.R: reconciled the stale "more than 130 productivity functions" blurb with the current true count (173, per the auto-generated function list header).
• Typos: matrix.fromVector ("Swap rows an columns" → "and"), split_vec_to_list_by_N ("A numeric oggr character vector" → "or").
• mean_normalize: @description said "median"; the code and its own diagnostic message both say "mean" — copy-paste drift from the neighboring median_normalize. Fixed the doc.
• as.list.df.by.row / as.list.df.by.col: their @description texts were swapped relative to actual behavior (by-row said "by its columns" and vice versa).
• symdiff: added the "not mathematically correct for >2 vectors" caveat (already printed at runtime) to the roxygen docs so they match.
• clip.outliers.at.percentile: the high param is documented as "Clip above threshold? Default: TRUE" but is never referenced in the function body — it always clips both tails. Git history (caf4d60) shows a previous PR deliberately reverted an attempt to make it do something, so per discussion with the repo owner we're keeping current both-tail-clipping behavior and just fixing the doc to say the param is currently unused/reserved.
• Dead code removal: a commented-out browser() call, a commented-out debug print, and two leftover interactive-session default arguments referencing undefined objects (UVI.assignment.filtered.3.HF, Sections.ls.Final) that only worked by accident (R's lazy argument evaluation never forced them).
• Stale tag: removed an unused @importFrom dplyr bind_rows (the only reference to bind_rows is a commented-out line; the function uses tibble::tibble()).

Bonus: 5 broken @examples found while verifying this PR

Running a full sweep of every man/*.Rd example (see Testing) surfaced 5 more documentation-only bugs — each is an example that doesn't reflect how the function actually needs to be called, not a behavior change:

• dput_pretty: example called the non-exported alias pretty_dput(vec) instead of the exported dput_pretty(vec) (the function is defined as dput_pretty <- pretty_dput <- function(vec) {...}, but only dput_pretty is exported).
• list2fullDF.byNames: example used unnamed vectors (list(set.1 = LETTERS[1:5], ...)), but the function requires named vectors as list elements — its own default argument value already uses vec.fromNames(...) correctly, so the example was updated to match.
• merge_1col_dfs_by_rn: example called the function with no arguments even though its only parameter, list_of_dfs, has no default — replaced with a minimal working example.
• splititsnames_byValues: example passed an unnamed vector into a function that requires !is.null(names(namedVec)).
• na.omit.mat: example called it directly as na.omit.mat(mat), but its name matches R's generic.class naming convention, so roxygen2/R register it as an S3 method for na.omit() instead of a normal export — it's only reachable via na.omit(mat) after class(mat) <- c("mat", class(mat)). Fixed the example to show that. (The underlying "exported-but-unreachable-under-its-own-name" API quirk is a real, if minor, functional issue — left as a follow-up backlog item since fixing it properly means renaming or force-exporting, a behavior change, not a doc fix.)
• Also added 3 missing @param entries (pFilter's v, pU's head_n, savehistory_2rstudio's history_file) that were causing an R CMD check WARNING.

All of the above were regenerated into NAMESPACE/man/*.Rd via roxygen2::roxygenise(".").

Testing

• R CMD build . + R CMD check on the built tarball: passes with only pre-existing, unrelated issues left untouched by this PR — confirmed identical before/after this diff:
  • S3 generic/method consistency WARNING for na.omit.mat (the same S3-naming quirk noted above).
  • A NOTE listing "no visible global function definition" for llprint, FirstCol2RowNames, qbarplot — these are the 3 undeclared cross-package dependency crashes already scoped for PRs 2-4 of this batch.
  • Missing base-package importFrom entries (abline, hist, is, head, tail, etc.) — long-standing, unrelated to text/docs, tracked as a separate backlog item.
• Additionally wrote a small script that extracts and runs every one of the 178 man/*.Rd examples individually (tools::Rd2ex + sys.source, package attached): 177/178 pass. The sole remaining failure is pU(), a pre-existing, already-known missing-@export bug (confirmed via git history) intentionally left for its own dedicated PR since exporting a previously-unreachable function is a behavior/API change, not a doc fix.
• lintr::lint_package(): reports ~1000 pre-existing style lints (line-length, commented-code) across the whole file. No .lintr config exists in the repo, and Development/Create_the_CodeAndRoll2_Package.R itself calls out a separate, future styler::style_pkg() pass as the intended path to compliance — this codebase predates lintr/tidyverse-style adoption. None of the lines this PR touches introduce a new category of lint not already present nearby.
• No testthat/automated tests added, per repo convention.

---

Generated by Claude Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4b8c64cd20

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".