test: migrate math/base/special/roundnf to ULP-based assertions

[Planeshifter]

Resolves a part of #11352.

Description

What is the purpose of this pull request?

This pull request:

• Migrates math/base/special/roundnf from manual ULP-difference tolerance checks to the canonical isAlmostSameValue-based assertion pattern described in [RFC]: Migrate math/base/special packages from relative tolerance testing to ULP difference testing (tracking issue) #11352.
• In test/test.js and test/test.native.js, replaces var ulpdiff = require( '@stdlib/number/float32/base/ulp-difference' ); plus manual delta/t.ok( delta <= N, ... ) checks with var isAlmostSameValue = require( '@stdlib/number/float32/base/assert/is-almost-same-value' ); and t.strictEqual( isAlmostSameValue( actual, expected, N ), true, 'returns expected value' );.
• Where the measured ULP delta was always 0 (i.e., the values were exactly equal), uses a plain t.strictEqual( actual, expected, 'returns expected value' ) instead, mirroring the existing convention used in other converted packages (e.g., sqrtf, atan2f).
• Tightened each ULP bound to the minimum integer that still passes the full fixture set:
  • "rounding a numeric value to a desired number of decimals": 0 ULP (exact) for all cases.
  • "rounding a numeric value ... can result in unexpected behavior": 0 ULP (exact).
  • "rounding a numeric value to a desired number of digits": 1 ULP for the two non-exact comparisons.
  • "rounding very small numbers (including subnormals)": 1 ULP (previously allowed up to 2 ULP).
• Also fixed a pre-existing, unrelated stdlib/doctest-marker ESLint violation on the touched comment line in test/test.js (// => → // returns), since it was blocking lint on a file this PR already modifies.

Related Issues

Does this pull request have any related issues?

This pull request has the following related issues:

• [RFC]: Migrate math/base/special packages from relative tolerance testing to ULP difference testing (tracking issue) #11352

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.

• test/test.native.js mirrors the same ULP bounds as test/test.js, following the pattern used in other converted packages (e.g., atan2f, sqrtf). The native addon could not be built/tested in this environment, so native-side bounds were not independently measured; they were kept identical to the JS-side measured bounds since no environment-specific divergence is expected for this simple rounding routine.
• Ran the full test suite for this package twice at the final ULP bounds to confirm determinism (340/340 passing both times).

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 helps reviewers determine how much scrutiny to apply when reviewing your contribution. Example disclosures: "This PR was written primarily by Claude Code." or "I consulted ChatGPT to understand the codebase, but the proposed changes were fully authored manually by myself.".

This PR was written by Claude Code, which discovered the candidate package, studied prior-art conversions (e.g., PR #11297, #14178, #12776), determined the minimum required ULP bound per assertion empirically by running the test suite, and applied the migration.

---

@stdlib-js/reviewers

---

🤖 Generated with Claude Code

https://claude.ai/code/session_01QQpKhCrYKm4SBFZgyCSduQ

---

Generated by Claude Code

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
math/base/special/roundnf	$\\color{red}271/273$ $\\color{green}+99.27\\%$	$\\color{red}22/23$ $\\color{green}+95.65\\%$	$\\color{green}2/2$ $\\color{green}+100.00\\%$	$\\color{red}271/273$ $\\color{green}+99.27\\%$

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