docs: follow-up fixes for commits merged to develop on 2026-05-26

[Planeshifter]

Description

Follow-up fixes for commits merged to develop between 2026-05-26 01:38 PT (f42a835) and 2026-05-26 03:32 PT (58bbb07). Three issues survived filtering across two independent style audits and two independent bug scans; each is grouped into its own per-package commit.

This pull request:

• Fix README tagline and Usage description in lib/node_modules/@stdlib/blas/base/ndarray/dsdot/README.md (introduced in 496a0da) to note that accumulation is performed in double precision, distinguishing dsdot from sdot and matching the existing repl.txt documentation.
• Align lib/node_modules/@stdlib/blas/ext/base/dxsa (introduced in ec4adc0) with sibling sxsa/gxsa conventions:
  • Rename mislabeled // VARIABLES // section header in lib/dxsa.js to // MODULES // to match the stdlib convention for require-only blocks.
  • Replace @returns input array with @returns `x` across all three signatures in docs/types/index.d.ts.
  • Repoint the [@stdlib/array/float64] link definition in README.md from the repo root to the package tree URL.
• Correct lib/node_modules/@stdlib/blas/base/ndarray/sdsdot/docs/repl.txt (introduced in e8b5180), which incorrectly stated the function returns 0.0 for empty input — the accumulator is initialised to the scalar third argument and returned directly when N <= 0, as the existing tests already asserted.

Related Issues

No.

Questions

No.

Other

The 24-hour window covered 18 first-parent commits to develop (f42a835..58bbb07), dominated by new blas/ext/base packages (caxpb, dxsa, sxsa, daxpb), new blas/base/ndarray wrappers (dsdot, sdsdot, scasum), the isMatrixTranspose→resolveStr refactor across the gemm family, and routine doc/namespace housekeeping.

Validation:

• Compared each new package against the established sibling reference (sxsa/gxsa for dxsa; ddot/dasum for the new ndarray wrappers; daxpb/saxpb for caxpb).
• Audited each commit's diff for code-style compliance against docs/style-guides/javascript.
• Scanned the diff for objective bugs (wrong arguments, off-by-one, contract mismatches between docs and code/tests).

Deliberately excluded:

• The isMatrixTranspose→resolveStr refactor pattern (deliberate API broadening to accept integer CBLAS enum codes).
• Phrasing-only differences in Notes/See-also sections across sibling packages where no canonical wording exists.
• Anything requiring code outside the 24-hour diff window to validate.

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: 24 hours of develop commits were reviewed by four parallel agents (two style audits, two bug scans); only findings that survived independent re-verification against the diff and the relevant sibling reference package are included.

---

@stdlib-js/reviewers

---

Generated by Claude Code

[stdlib-bot]

Coverage Report

No coverage information available.

[kgryte]

As the CI failure is not due to changes introduced in this PR, I'll go ahead and merge.