docs: backfill W1.5 board entries and correct the stale W1.5 contract section

[AdaWorldAPI]

Summary

Two hygiene gaps found while closing out the #293/#294/#295 arc. Markdown only — no source, tests, or build files touched.

1. The board had no record of any of it

.claude/blackboard.md and .claude/board/AGENT_LOG.md contained zero mentions of the W1.5 signature primitives. All three PRs landed with no board entry, against CLAUDE.md's Agent Protocol ("write decisions to the blackboard, not just to chat"). A session reading the blackboard would not have known signature_pde or randomized_signature existed at all.

Entries added for #293, #294 and #295 carrying the decisions, the measured numbers, and the loose ends — including the two that matter most: the debug_assert_eq!-compiles-out-in-release bug #294's follow-up fixed, and the caveat that #295's 100% docstring figure is a manual /// scan rather than bot-confirmed (it merged ~5s after leaving draft, so CodeRabbit never re-ran).

2. The W1.5 contract section was describing a world that no longer exists

.claude/knowledge/vertical-simd-consumer-contract.md still presented all three W1.5 primitives as future work "gated on jc Pillar 11 activation". That gate opened 2026-05-07 (lance-graph:crates/jc/src/lib.rs:26), and #6 and #7 have shipped.

Regraded in place using the workspace's existing ⊘-blockquote convention — original sketches kept verbatim, each followed by a correction block, rather than silently rewritten.

The correction worth carrying forward

This doc's API sketches have now been wrong the same way twice. Both #6 and #7 were sketched as f32/F32x16 while the real consumer (lance-graph crates/sigker) is f64/Vec<f64> — both shipped on F64x8. #294 additionally found the sketch wrong about buffer ownership (projections are materialized once per encoder and reused, not re-derived per step) and about shape (k is runtime, so the hot path is a k×k GEMV, not a single-register update).

W1.5-#8's I16x16 sketch is therefore marked UNVERIFIED — a 2-for-2 miss rate — and must be read off crates/sigker/src/log_signature.rs before any implementation starts.

A companion scoping pass this session confirmed the third miss and went further: log_signature.rs:272 is Vec<f64>, so I16x16 is the wrong element width and the wrong signedness family. Findings are recorded on the lance-graph board rather than here, to keep the separation of concerns.

Also recorded: W1.5-#8 is unblocked, not deferred.

Test plan

• Markdown only — no compiled surface changed, so cargo state is unaffected by this diff
• Board files verified append-only (insertions only, zero deletions)
• ⊘ correction convention matched against existing usage in the same doc family

🤖 Generated with Claude Code

https://claude.ai/code/session_016WkNBjHc2e3zuyz9i8qJEv

---

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_34f1c47b-c2b9-40e2-b87c-b3a672b98012)