bench: add missing native benchmark

[Planeshifter]

Description

Adds the missing benchmark/benchmark.native.js to @stdlib/math/base/special/log1pmx so its N-API addon has the same JS-level benchmark surface as its 342 siblings in @stdlib/math/base/special.

Namespace summary

• Namespace: @stdlib/math/base/special
• Members analyzed: 357 packages, none autogenerated.
• Native subset: 343 packages ship a native N-API addon (binding.gyp + src/addon.c + lib/native.js); 14 are pure-JS by design.
• Features with a clear majority (≥75%): the shared file tree (README.md, lib/index.js, lib/main.js, docs/repl.txt, docs/types/{index.d.ts,test.ts}, test/test.js, benchmark/benchmark.js, examples/index.js, package.json), the standard package.json key set, and the H2 section list (Usage, Examples, C APIs for native packages). Among the 343 native packages, benchmark/benchmark.native.js and test/test.native.js are each present at 342/343.
• Features excluded (no clear majority): the stdlib package.json sub-key set (only 249/357 packages carry a __stdlib__ block, below the 75% cutoff — likely a build-time addition, deliberately not touched here); the H2 See Also section (237/357, generator-owned per the routine's auto-populated gate).

math/base/special/log1pmx

log1pmx is the sole native package in the namespace missing benchmark/benchmark.native.js — 342 of 343 native siblings (99.7%) carry it, including the immediate sibling log1p. The package already ships a functional N-API addon (binding.gyp, src/{addon.c,main.c}, lib/native.js), so the gap is a bookkeeping oversight from when the package was added, not an intentional deviation. The new file mirrors log1p/benchmark/benchmark.native.js verbatim in structure — loads the addon via tryRequire('./../lib/native.js'), skips cleanly when the addon isn't built, and reuses the JS benchmark's uniform( 100, -1.0, 99.0 ) input distribution so JS and native numbers are comparable. No other file in the package changes, and nothing in test/, examples/, README.md, or docs/repl.txt — nor anywhere else in the repo — references or depends on the file's absence.

Related Issues

None.

Questions

No.

Other

Validation

• Structural extraction over all 357 members — file trees, package.json key sets, manifest.json key sets, and README H2/H3 section lists — computed the majority pattern per feature at the routine's 75% threshold.
• Semantic extraction via per-package agents was skipped for this run because structural drift alone surfaced a clear, single-file correction; the namespace is dense enough (357 members) that a per-package semantic pass would have burned rate limits for no gain against the specific finding already in hand.
• Three-agent drift validation (opus semantic-review, opus cross-reference, sonnet structural-review) each independently returned confirmed-drift for the log1pmx finding.
• Cross-run dedup against open PRs on the namespace: PR test: add special value tests for log1pmx #9491 modifies log1pmx/test/test.js (adding special-value tests) and PR docs: fix TSDoc issues across many math/base/special declarations #12467 revises TSDoc across math/base/special; neither touches benchmark/benchmark.native.js in log1pmx and neither collides at the file level.

Deliberately excluded:

• log1pmx/test/test.native.js (also missing, 342/343 conformance). Adding it would mirror test/test.js, which PR test: add special value tests for log1pmx #9491 is actively extending; landing a test.native.js derived from today's test.js would silently omit the special-value tests once test: add special value tests for log1pmx #9491 merges. Deferred until test: add special value tests for log1pmx #9491 lands.
• cfloor/benchmark/c/native/{Makefile,benchmark.c} (native package, 340/343 conformance for this pair). cfloor's existing benchmark/c/benchmark.c is a raw-C reference (calls floor() on creal(z)/cimag(z)); adding a stdlib-impl native benchmark would require writing ~140 lines of C that construct stdlib_complex128_t inputs and format Complex128 outputs — beyond the routine's mechanical-fix criterion.
• hyp2f1/benchmark/c/native/{Makefile,benchmark.c} and sqrtpif/benchmark/c/native/{Makefile,benchmark.c} (same 340/343 conformance). Both packages have benchmark/c/benchmark.c files that already #include "stdlib/math/base/special/*.h" and call stdlib_base_* — content-wise they belong under benchmark/c/native/. The correction is a directory move plus Makefile CFLAGS rework, not a pure file addition, so it needs a human call on whether to move or duplicate; dropped from this run.
• The fast sub-namespace parent (single-outlier missing lib/main.js, benchmark/benchmark.js, docs/repl.txt). It is a namespace package, not a leaf; deviation is intentional.
• Autogenerated ## See Also sections (237/357), per the routine's auto-populated gate.

Checklist

• Read, understood, and followed the contributing guidelines.

AI Assistance

• Yes
• No

If you answered "yes" above, how did you use AI assistance?

• Code generation (e.g., when writing an implementation or fixing a bug)
• Test/benchmark generation
• Documentation (including examples)
• Research and understanding

Disclosure

This PR was authored by Claude Code on behalf of @Planeshifter as a cross-package drift audit of @stdlib/math/base/special. Structural features (file trees, package.json shape, README sections, manifest shape) were extracted for all 357 members; majority patterns were computed at the routine's 75% threshold; the single surviving outlier finding was independently validated by three reviewer agents (opus semantic-review, opus cross-reference, sonnet structural-review). A human will audit and promote the PR out of draft.

---

@stdlib-js/reviewers

---

Generated by Claude Code

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
math/base/special/log1pmx	$\\color{green}207/207$ $\\color{green}+100.00\\%$	$\\color{green}12/12$ $\\color{green}+100.00\\%$	$\\color{green}3/3$ $\\color{green}+100.00\\%$	$\\color{green}207/207$ $\\color{green}+100.00\\%$

The above coverage report was generated for the changes in this PR.