wgsl: Convert quantizeToF16 to used hfround

[zoddicus]

Instead of passing the input through a F16Array, use the library provided function hfround. hfround is a fast look up table based rounding function for f16.

Benchmarking locally this provides a ~20% improvement to fma interval calculations, which are particularly sensitive to quantization cost. Overall I was seeing more on the order of ~10% improvement.

---

Requirements for PR author:

• All missing test coverage is tracked with "TODO" or .unimplemented().
• New helpers are /** documented */ and new helper files are found in helper_index.txt.
• Test behaves as expected in a WebGPU implementation. (If not passing, explain above.)

Requirements for reviewer sign-off:

• Tests are properly located in the test tree.
• Test descriptions allow a reader to "read only the test plans and evaluate coverage completeness", and accurately reflect the test code.
• Tests provide complete coverage (including validation control cases). Missing coverage MUST be covered by TODOs.
• Helpers and types promote readability and maintainability.

When landing this PR, be sure to make any necessary issue status updates.

[zoddicus]

There is probably a similar optimization to be had from using Math.fround in quantizeToF32