chore: correct issues found during recent develop commit review

[Planeshifter]

Follow-up fixes for commits merged to develop between 2026-05-18 16:46:06 -0700 and 2026-05-19 04:22:25 -0700.

Description

What is the purpose of this pull request?

This pull request applies follow-up corrections for validated, high-signal issues identified while reviewing the 13 commits merged to develop in the window above (7fab4532...e1d7dc74).

That window covered: two new BLAS extension packages (blas/ext/base/daxpb, blas/ext/base/saxpb) with their namespace and TypeScript wiring; a blas/base/cgemv bug-and-typo fix; added α/β tests for blas/base/ggemm; doc/grammar propagation across the gindex-of-column/sindex-of-column siblings; an ndarray/base benchmark refactor to string interpolation (~200 files); and an .npmrc configuration update.

This pull request fixes the following, grouped by package. Each group is a separate commit (fix: for the configuration bugs, style: for the JSDoc tag removal):

.npmrc

• commit-hooks was misspelled as commits-hooks (e1d7dc74). npm silently ignores the unrecognized key, so git commit hooks would not run during npm version. Corrected the spelling.
• min-release-age used a colon separator — min-release-age: 90 (e1d7dc74). .npmrc is INI-format and requires =; every other directive uses it, and with a colon npm does not parse the directive, so the minimum release-age supply-chain guard never takes effect. Changed to min-release-age = 90.

blas/ext/base/saxpb

• lib/saxpb.native.js (103392c9) carried a spurious @private JSDoc tag on the exported native function — a public entry point, not an internal helper. Removed the tag to match the sibling daxpb package and the dapx reference.
• lib/ndarray.native.js (103392c9) likewise tagged the exported native ndarray interface @private. Removed the tag to match daxpb and dapx.

Related Issues

Does this pull request have any related issues?

None.

Questions

Any questions for reviewers of this pull request?

No.

Other

Any other information relevant to this pull request? This may include screenshots, references, and/or implementation notes.

This pull request is the output of an automated review of the last 24 hours of develop. Four reviewer agents — two for stdlib code-style compliance, two for bugs/correctness — independently audited the window's diff; findings were cross-checked, and only issues re-verifiable directly against the diff (and the relevant stdlib reference packages) were retained.

Validation — checked, no action required:

• blas/base/cgemv (c606f03c): the LDA < v → LDA < vala correction, the cfill_ndarray zero-argument fix, and the strideA2 JSDoc/grammar fixes are all genuine and correct.
• blas/base/ggemm tests (fc3275c0): the new fixtures' C_out values verified arithmetically (α*A*B + β*C).
• daxpb/saxpb: README, REPL, and JSDoc numeric examples verified (x[i] = α*x[i] + β); C sources, addons, and manifests internally consistent.
• ndarray/base benchmark refactor (7fab4532, 9330de70): the format() conversions preserve the original concatenated strings; format-specifier counts match argument counts.

Deliberately excluded (out of scope / requires interpretation): the daxpb/saxpb ndarray JSDoc descriptions differ slightly in whether they include the phrase "using alternative indexing semantics" — a wording inconsistency with no governing style-guide rule, left for maintainer judgment. Auto-generated bot commits (REPL data, namespace ToC, TypeScript declarations) were not line-reviewed.

Checklist

Please ensure the following tasks are completed before submitting this pull request.

• Read, understood, and followed the contributing guidelines.

AI Assistance

When authoring the changes proposed in this PR, did you use any kind of 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

If you answered "yes" to using AI assistance, please provide a short disclosure indicating how you used AI assistance.

This PR was generated by an automated multi-agent review of commits merged to develop over the preceding 24 hours, run with Claude Code. Reviewer agents flagged candidate issues; each retained fix was independently verified against the diff and the relevant stdlib reference packages before being applied. This is a draft — a human maintainer should audit before promoting it.

---

---

Generated by Claude Code

[kgryte]

In the CI failure, the addon successfully built locally, so I will go ahead and merge this PR, as the failure does not appear to be related to changes introduced here.