fix: evaluate large arguments accurately in math/base/special/fresnelf

[Abhist17]

Description

What is the purpose of this pull request?

This pull request:

• applies fix: evaluate large arguments accurately in math/base/special/fresnel, fresnelc and fresnels #16118 to the single-precision math/base/special/fresnelf (JS and C).

It has the same two problems, and the second is worse in single precision:

1. For x > 36974 it returned exactly 0.5, although the oscillating terms are of size 1/(πx).
2. sincosf( HALF_PI * x2 ) used the rounded x*x. A float32 x*x near x = 3000 is off by up to 0.5, which is most of a radian of argument.

The fix reduces x² modulo 4 exactly. A Veltkamp split with 4097 = 2^12+1 leaves 12-bit halves, so every partial product is exact in single precision, and each one is reduced with fmodf. The asymptotic branch now runs up to 2^24, where 1/(πx) is below half an ulp of 0.5.

• replaces the C/Cephes fixtures with the BigFloat Julia runner from fix: evaluate large arguments accurately in math/base/special/fresnel, fresnelc and fresnels #16118, evaluated at single-precision inputs (huge.json now runs log-spaced from 4e4 to 2^24). Against those exact values:

set	develop	this PR (S / C)
small [1e-4, 9.99]	20 ULPs	3 / 7
medium [10, 100]	160	1
large [100, 36968]	5120	0
huge [4e4, 2^24]	320	1

Related Issues

Does this pull request have any related issues?

None. Follows #16118.

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.js and test.native.js pass.

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

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
math/base/special/fresnelf	$\\color{red}574/578$ $\\color{green}+99.31\\%$	$\\color{red}28/30$ $\\color{green}+93.33\\%$	$\\color{green}7/7$ $\\color{green}+100.00\\%$	$\\color{red}574/578$ $\\color{green}+99.31\\%$

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