toolchain: Rust 1.97.1 → 1.98.1, and the chunks_exact → as_chunks shift it forces

[AdaWorldAPI]

What

Workspace-wide sweep to current stable (1.98.0 released 2026-08-20, 1.98.1 on 2026-09-01). Repo 2 of 9. This repo's own rule is honoured: the channel and Cargo.toml's rust-version move together (1.97 → 1.98).

Two findings, where lance-graph had one

finding	sites
clippy::chunks_exact_to_as_chunks — new in 1.98, default-on style group	6
clippy::from_iter_instead_of_collect — removed in 1.98, and carried in src/lib.rs's crate-level allow list, so the allow itself is now an error	1

The second one does not exist in lance-graph, which is why the sweep measures each repo rather than generalising from the first.

Why as_chunks is the right shape here, not just lint appeasement

chunks_exact was the correct call when these buffers were 4096 and 16384 wide: the remainder is always empty at those widths and the runtime width check is dead weight. At the 32+96 shape, the classid plus the twelve-byte facet payload, the width is a compile-time fact, and as_chunks::<N> returns &[T; N] — a type that carries the width instead of re-checking it. That is the honest representation for a fixed-width register.

Two of the six wanted the destructuring form rather than a mechanical swap: asum_f32 and asum_f64 called chunks_exact(N) and chunks_exact(N).remainder() separately, and as_chunks returns both halves at once, so each kernel now makes one call where it made two.

The toolchain comment is rewritten, not edited

It opened # Pinned to 1.97.1, restating the version in prose beside the channel line. This file already carries the warning for exactly that failure: "A stale comment on a version pin is how the next reader learns the wrong number." A bump edits channel and leaves prose behind, so the prose now keeps only the rationale, plus an append-only bump log.

Gate

These are AVX-512 kernels, so the test suite is the real gate, not clippy:

• cargo clippy --all-targets -- -D warnings → exit 0
• cargo fmt --all -- --check → clean
• cargo test --lib → 2278 passed, 0 failed, 30 ignored

All run through bare cargo on the new pin, which is what CI does.

🤖 Generated with Claude Code

https://claude.ai/code/session_01AGVLyRZNEKKBSfBDJfbY3V

---

Generated by Claude Code

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

---

_{Comment @coderabbitai help to get the list of available commands.}

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_bec01a95-e227-4cc7-bff3-15a90c848942)

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.