impl(sprint-11/wave-A): D-CSV-1 + D-CSV-3 + D-CSV-4 — causal-edge v2 + signed-mantissa NARS + CollapseGateEmission

.claude/board/AGENT_LOG.md

@@ -1,3 +1,28 @@
+## [Fleet sprint-11-wave-a-impl] [IN PR] D-CSV-1 + D-CSV-3 + D-CSV-4 (branch claude/sprint-11-wave-a-impl, commit ab39d01)
+
+**D-id(s):** D-CSV-1 (causal-edge v2 layout), D-CSV-3 (signed-mantissa InferenceType expansion), D-CSV-4 (CollapseGateEmission in contract).
+
+**Workers (2 Sonnet, parallel):**
+- **W-A1** — D-CSV-1 + D-CSV-3 paired in causal-edge crate. NEW `layout.rs` (~130 LOC, all shift constants + masks + TrustTexture + compile-time _LAYOUT_COVERAGE const-assert); EXTEND `edge.rs` with v2 accessors (inference_mantissa i4-signed, w_slot, truth, spare, with_routing(w,t) — no G-slot); NEW `v2_layout_tests.rs` (16 tests covering signed-mantissa round-trip, field-isolation matrix, 2-arg with_routing, spare isolation, size_of==8). Cargo bumped 0.1.0 → 0.2.0 with `default = ["causal-edge-v2-layout"]`. `InferenceType::to_mantissa/from_mantissa` provides bidirectional v2 mapping while keeping the enum intact for v1 callers.
+- **W-A2** — D-CSV-4 in contract crate. NEW `MailboxId = u32` + `CollapseGateEmission` (Vec instead of SmallVec to preserve contract zero-dep, with documented deferral to sprint-12+ optimization). API: new/push_baton/baton_count/wire_cost_bytes (13 + 10×N) + provenance accessors. 8 tests pass.
+
+**Main-thread P0 caught in code review:** worker W-A1 left v1 `pack()` writing `temporal << 52` even under v2 feature, corrupting the new reclaim zone (bit 52 = plasticity[2], 53-58 = W, 59-60 = lens, 61-63 = spare). Same root cause as the W3 spec codex P1 from PR #381. Fixed by feature-gating the temporal write in pack() so v2 silently drops the arg; two v1-only tests (`test_roundtrip`, `test_temporal_in_msb_gives_sort_order`) gated on `#[cfg(not(feature = "causal-edge-v2-layout"))]`.
+
+**OQ ratifications absorbed:** OQ-CSV-2 = 6 bits (default per plan §11 recommendation). OQ-CSV-1 + OQ-CSV-4 deferred to Wave B (D-CSV-2 / D-CSV-5).
+
+**Test status:**
+- causal-edge v2 (default): 30 pass / 1 fail (test_build_fast — pre-existing on main, confirmed via stash-revert)
+- causal-edge v1 (no default features): 16 pass / 1 fail (same pre-existing)
+- lance-graph-contract collapse_gate: 8/8 pass
+- lance-graph-planner: compiles with 2 deprecation warnings (`inference_type()`, `temporal()`) — the intended migration signal for downstream callers
+- p64-bridge: compiles with 1 deprecation warning
+
+**Outcome:** Sprint-11 Wave A scope (Phase A substrate primitives) reaching merge gate. Wave B (D-CSV-2 QualiaI4_16D + D-CSV-5 QualiaColumn migration) blocked on OQ-CSV-1 (qualia 16D per-dim assignment) — needs qualia-engineer agent cross-check before spawn.
+
+**Pre-existing finding:** `tables::tests::test_build_fast` fails on clean main under both feature configurations; not introduced by this PR. To be filed in ISSUES.md separately.
+
+---
+
 ## [Fleet sprint-log-csv-prep] [DONE] cognitive-substrate-convergence-v1 spec patches (PR #381 merged 2026-05-16)
 
 **D-id(s):** Pre-sprint-11 spec-patch bundle for all 8 active D-CSV-* deliverables (D-CSV-1..D-CSV-7, D-CSV-11) — patches to 8 sprint-10 specs without implementation; sprint-11 implementation spawn still gated on user ratifications.

.claude/board/STATUS_BOARD.md

@@ -426,10 +426,10 @@ Consolidates sprint-10 architectural decisions before context dilution.
 
 | D-id | Title | Status | PR / Evidence |
 |---|---|---|---|
-| D-CSV-1 | `causal-edge` crate v2 layout (signed mantissa, W-slot, lens, drop temporal) | **Queued** | blocked on OQ-CSV-1 + OQ-CSV-2 user ratification |
+| D-CSV-1 | `causal-edge` crate v2 layout (signed mantissa, W-slot, lens, drop temporal) | **In PR** | branch `claude/sprint-11-wave-a-impl`, commit `ab39d01`; OQ-CSV-2 ratified to 6 bits (default) |
 | D-CSV-2 | `QualiaI4_16D` type in `lance-graph-contract::qualia` + f32↔i4 migration helpers | **Queued** | blocked on OQ-CSV-1 (per-dim assignment) |
-| D-CSV-3 | InferenceType signed-mantissa expansion (absorbs PR-LL-1 Intervention/Counterfactual into canonical edge enum) | **Queued** | depends on D-CSV-1 |
-| D-CSV-4 | `CollapseGateEmission` wire format spec + impl per plan §8 | **Queued** | depends on D-CSV-1 |
+| D-CSV-3 | InferenceType signed-mantissa expansion (absorbs PR-LL-1 Intervention/Counterfactual into canonical edge enum) | **In PR** | branch `claude/sprint-11-wave-a-impl`, paired with D-CSV-1 in same crate |
+| D-CSV-4 | `CollapseGateEmission` wire format spec + impl per plan §8 | **In PR** | branch `claude/sprint-11-wave-a-impl`, contract crate (Vec instead of SmallVec to preserve zero-dep) |
 
 ### Phase B — Storage & dispatch path (sprint-11)
 

.claude/knowledge/woa-routing/README.md

@@ -0,0 +1,72 @@
+# WoA Flask Routing Table (Canonical Mirror)
+
+> **Canonical source:** [`AdaWorldAPI/WoA` `.claude/reference/routing_table.{json,csv}`](https://github.com/AdaWorldAPI/WoA/tree/main/.claude/reference) — generated by [PR #33](https://github.com/AdaWorldAPI/WoA/pull/33) (merged 2026-05-16).
+> **This is a mirror** for cross-repo greppability from lance-graph sessions — refresh by re-fetching from the canonical path.
+> **Routes:** 514 endpoints across the WoA app.py monolith + 53 blueprint modules under `woa/blueprints/*.py`.
+> **Refreshed:** 2026-05-16.
+
+## Schema
+
+`routing-table.json`:
+
+```json
+{
+  "generated_by": "...",
+  "generated_at": "...",
+  "route_count": 514,
+  "routes": [
+    {
+      "endpoint": "system_pages.index",
+      "function": "index",
+      "path": "/",
+      "methods": "GET",
+      "source_file": "woa/blueprints/system_pages.py",
+      "line": 19,
+      "blueprint": "system_pages",
+      "cluster": "system_pages"
+    },
+    ...
+  ]
+}
+```
+
+`routing-table.csv`: same fields as flat columns.
+
+## How WoA harvested it (per PR #33)
+
+> Walked every `*.py` under:
+> - `app.py` (91 routes still in the monolith)
+> - `woa/blueprints/*.py` (423 routes across 53 files)
+> - `woa/*.py` and `woa_*.py` (root-level helpers)
+>
+> For each `FunctionDef`, scanned `decorator_list` for any `@.route(...)` call, pulled path + methods, joined with the blueprint's `url_prefix`.
+>
+> Coverage: 514 of 515 runtime routes captured; the delta is Flask's auto `/static/` rule (no decorator → invisible to AST).
+
+## When to refresh
+
+When PR #33's harvester runs again on a newer WoA HEAD (e.g. after further refactors moving more routes out of the monolith), re-fetch from the canonical path. Drop a dated note in this README on each refresh.
+
+The local fetch pattern (used by the 2026-05-16 refresh):
+
+```python
+# via mcp__github__get_file_contents on AdaWorldAPI/WoA path
+# .claude/reference/routing_table.csv (and .json), strip the
+# "[Resource from github at ...]" wrapper prefix, write the body
+```
+
+## Use cases (cross-repo, from a lance-graph session)
+
+- **Plan further extractions:** filter `cluster == "app.py-monolith"` to see what's left in the 91-route remnant.
+- **Verify no route disappears across refactors:** diff `route_count` before/after.
+- **Membrane wiring (callcenter / sharepoint / spear / q2):** identify which WoA endpoints feed which downstream consumer faster than grepping the multi-file blueprint tree.
+- **Sprint planning** for lance-graph ↔ WoA integration: this table is the contract surface.
+
+## What this is NOT
+
+- Not a live runtime extraction (no `app.url_map` introspection). Pure static AST per PR #33's harvester. Dynamic decorator expressions (computed `route(some_var, ...)`) are skipped silently.
+- Not a security/auth model. The harvester does NOT include the auth/csrf/limiter decorator chain (unlike the lance-graph session's earlier partial 96-route harvest of just `app.py` — superseded by this mirror).
+
+## History
+
+- **2026-05-16** initial harvest in this lance-graph session against `WoA/app.py` only — 96 routes from the monolith. Superseded the same day by this canonical mirror from WoA PR #33, which covers all 514 routes across the modular blueprint architecture.