Check Zig formatting on a pull request

[AdrianBinDC]

What this changes

zig fmt on zig-flashnext still wants 22 files changed. This formats those files and adds a pull-request check that runs zig fmt --check build.zig zig. No behavior change. Ruff is not in this pull request. Issue #461 stays open until that cleanup lands too.

Receipt

• Environment, with the TensorFold commit, MLX or PyTorch version, machine and GPU, checkpoint and revision: base zig-flashnext at db28187, this commit on top. Zig 0.17.0. No model and no GPU work.
• Exactness, from tools/bench_concurrent.py --alone --serial or token_sha against "draft": false: not run. The Zig diff is line breaks from zig fmt. The new file is the formatting check.
• Decode speed before and after, from tools/bench_openai.py: not run.
• Prompt speed before and after, from tools/prefill_cold.py: not run.
• Tests run, with pass, fail and skip counts: zig fmt --check build.zig zig passes. zig build test on these Zig sources: 19/19 steps succeeded, 13/13 tests passed.
• Not run, and why: Metal and CUDA model runs, because no arithmetic changed. ruff check, on purpose: 1372 findings, and this pull request is the Zig half of Lint in CI after zig fmt and ruff are clean #461.

Checklist

• Every token still goes through the lane rounds. No serial path, and nothing that needs drafts off.
• Drafted output equals "draft": false, a resumed prompt equals a fresh one, and concurrent equals solo.
• No precision traded for speed. If the bits change, the description says which sums change and why.
• Prompt processing is no slower than the last release.
• The description says which platforms I ran, Metal M1 to M5 and CUDA, and which I could not.
• New tests fail before the change, pass after it, and skip cleanly without their dependency.
• Comments and docstrings are one line. No measurements or history in the source.
• No personal data, machine names, internal hosts or local paths. No AI attribution lines.