feat: add TypeScript declarations for blas/base/igamax

[Planeshifter]

Description

Adds the missing docs/types/index.d.ts and docs/types/test.ts files for @stdlib/blas/base/igamax. The package's package.json already declares "types": "./docs/types" — pointing at a directory that did not exist. This patch makes that pointer resolve.

Namespace summary

• Target: @stdlib/blas/base
• Members: 110 total; 107 voted (3 namespace-aggregator packages — assert, ndarray, wasm — excluded).
• Features extracted: file tree, package.json shape (top-level keys, scripts, keywords), manifest.json shape, README H2/H3 sections, test/benchmark/example file naming, plus a fast cross-package scan for error-construction patterns in lib/main.js.
• Features with a clear majority (≥75%): top-level package.json keys (100%), manifest.json shape where present (100%), lib/, test/, examples/, docs/ directories (100%), benchmark/ directory (98%), docs/types/ directory (99%), docs/repl.txt (98%), test/test.js (100%), examples/index.js (100%), benchmark/benchmark.js (97%), README ## Usage / ## Examples (100% / 98%), and error-construction (the 4 main files that throw all use format).
• Features without a clear majority (excluded): ## Notes (79% — many small enum/resolver utilities legitimately do not need notes), binding.gyp / include.gypi / src/ / native-bindings cluster (~47-53% — split intentionally between native and pure-JS implementations), ## C APIs (64%), ## See Also (23%), ## References (7%).

blas/base/igamax

igamax is the lone outlier on the docs/types/ feature (106/107 = 99% conformance). The package's own package.json declares "types": "./docs/types", but the directory does not exist; every other generic g* sibling (gasum, gaxpy, gcopy, gdot, ggemm, ggemv, gger, gnrm2, gscal, gswap, gsyr) ships one. The added index.d.ts adopts gasum's generic input typing — NumericArray | Collection<number> | AccessorArrayLike<number> — because igamax accepts any numeric array. The signature surface mirrors the existing idamax / isamax declarations (the closest i*amax siblings): a Routine interface with the 3-arg main overload and the 4-arg .ndarray overload, both returning number. The test.ts exercises both overloads against Float64Array and AccessorArray, with $ExpectError coverage for argument-type and arity mismatches matching the sibling conventions.

Validation

Checked:

• Structural feature extraction across all 110 members (3 aggregators excluded).
• Per-package agent extraction was scoped to error-construction patterns rather than full signature/JSDoc shape, since the 100% format-conformance signal made the wider semantic sweep unnecessary.
• Three-agent drift validation on the surviving candidate. Both reviewers (cross-reference, structural) returned confirmed-drift.
• Pre-validation open-PR collision check via GitHub MCP — no open PRs touch blas/base/igamax.
• No tests or examples in blas/base/igamax/ reference the absence of docs/types/; the README's documented signatures (igamax( N, x, strideX ) and igamax.ndarray( N, x, strideX, offsetX ), both returning an index) match the declarations exactly.

Deliberately excluded:

• blas/base/drotg and blas/base/srotg (missing ## Examples H2 inside the <section class="examples"> wrapper) — dropped on open-PR collision (PRs feat: add blas/base/zrotg #9628, feat: add C and Fortran implementation of blas/base/drotg #9592, feat: add C / Fortran implementation for blas/base/drotg #2264, feat: adds Fortran and C implementation of blas/base/srotg #9537, feat: add C and Fortran implementation for blas/base/srotg #987 all touch these packages).
• blas/base/dgemm, sgemm, ggemm (missing benchmark/benchmark.js) — intentional deviation: each ships ~40 parameter-variant benchmarks instead of a canonical entry; the structure reflects matrix-matrix multiply's parameter space, not drift.
• ## Notes outliers (22 packages, most enum/resolve utilities) — feature falls just above the 75% threshold and the outlier cohort is a coherent semantic subset, not drift.

Related Issues

None.

Questions

No.

Other

Generated by an automated cross-package drift-detection run (random-pick namespace selection; full per-feature distribution and dropped-candidate log live in the local report). The diff is mechanical: two new files, no edits to existing code, no observable behavior change.

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 as part of an automated cross-package drift-detection routine. The drift was identified by structural feature extraction across @stdlib/blas/base/; the proposed fix was validated by parallel cross-reference and structural-review subagents (both confirmed-drift); the .d.ts and .ts content was generated by modeling on the existing blas/base/gasum/docs/types/ (generic input typing) and blas/base/idamax/docs/types/ (index-returning Routine interface) siblings.

---

@stdlib-js/reviewers

---

Generated by Claude Code

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
blas/base/igamax	$\color{green}268/268$ $\color{green}+100.00\%$	$\color{green}19/19$ $\color{green}+100.00\%$	$\color{green}3/3$ $\color{green}+100.00\%$	$\color{green}268/268$ $\color{green}+100.00\%$

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