refactor: replace 1:length(x) with seq_along()/seq_len() sweep

[vertesy]

Motivation

Per the repo's own convention (integer-literal ranges like 1L:x should be avoided), this replaces every remaining 1:length(x) idiom with seq_along(x) (or seq_len(2L * length(x)) for the one doubled-length loop bound) across the file. Pooled into a single PR since it's the exact same mechanical substitution applied identically everywhere — per the workflow guideline, "replacing the same code convention everywhere → one PR."

What changed

13 functions, 15 call sites: translate, sortbyitsnames, split_vec_to_list_by_N, zigzagger, colsplit, rowsplit, merge_dfs_by_rn, merge_1col_dfs_by_rn, symdiff.ls, reorder.list, intermingle2lists, as.listalike, imovingSEM, and dput_pretty (two occurrences).

Bonus: also fixes a real edge-case bug

1:length(x) evaluates to the reversed range c(1, 0) when x has length 0 (since length(x) is 0 and 1:0 counts down, not up). This previously crashed sortbyitsnames() on empty input with a names-length mismatch. Verified:

sortbyitsnames(setNames(numeric(0), character(0)))
# before: Error in names<-(...) : 'names' attribute [2] must be the same length as the vector [0]
# after:  named numeric(0)

Two things deliberately left alone

• intermingle.cbind()'s two 1:(2 * length(df1)) occurrences — open PR intermingle.cbind: input validation, row-name alignment, safe subsetting, loop fix, docs #38 (codex/update-intermingle.cbind-function) already fixes those as part of a larger, unrelated hardening change to that function (input validation, row-name alignment, safe subsetting). Touching them here would just create a merge conflict with no benefit — left for that PR.
• One occurrence inside a comment string in as.listalike() (# Iterate over the list, and fill in the elements with the corresponding elements from the vectorfor (v in 1:length(list_wannabe)) {) — this comment looks corrupted, as if merged with a stray copy of the code on the next line. Per the repo's own convention ("if a comment's purpose is unclear or looks outdated, flag it rather than deleting or rewriting it — I'll decide"), I left it as-is and only fixed the actual code line below it. Happy to clean it up if you'd like — just say what it should say.

One related, narrower limitation noted (not fixed here)

colsplit()/rowsplit() still fail on a fully zero-length factor input (0 unique levels), but for a different, deeper reason unrelated to this style sweep: when the loop never runs, names(ListOfDFs) <- levelz tries names(NULL) <- character(0), which itself errors in base R regardless of how the loop bound is written. That's a separate, minor, pre-existing limitation — worth a follow-up if this edge case matters, not folded into this mechanical PR.

Testing

• R CMD build + R CMD check: no new NOTEs/WARNINGs/ERRORs vs. the unmodified dev branch — same pre-existing, unrelated issues only (already tracked in the other PRs in this batch).
• Re-ran every one of the 178 man/*.Rd examples individually (tools::Rd2ex + sys.source): identical pass/fail set as unmodified dev (7 pre-existing failures, all already tracked elsewhere in this batch) — no new failures introduced.
• Spot-checked normal usage of every touched function (translate, zigzagger, dput_pretty, reorder.list, intermingle2lists, imovingSEM, etc.) to confirm unchanged behavior for non-empty input.
• 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: 9c12b7958c

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".