feat: graph context injection — the session's prior knowledge, steered in at session start (vision-context-injection.md)

One shot per chat, fired at the first turn's onFinish: the session's
topic (its only typed message) is embedded as an EmbeddingGemma retrieval
query; the semantically closest graph propositions are selected under the
session's token budget (pure budget, no similarity threshold — the
decided doctrine) and delivered as a marker-only steer through the wake
pipeline, queuing behind pending questions like every sub-agent result.

- contracts: ChatSettings.graphContextTokens (default 1024, max 4096,
  step 128, 0 = off); GraphContextData + graphContextText; the
  graph-context JobKind + buildGraphContextSteer (id steer_ctx_<chatId>)
- core/graph-context.ts: brute-force cosine top-K, greedy chars/4 fill
  under budget, most-recent UnderstandingLevel fold, count cap 64,
  parse-don't-cast at the marker seam
- server: embedQuery (role-asymmetric query format, same warm singleton
  — vector-space parity by construction); the runner (startGraphContext-
  Injection one-shot gates; runGraphContextJob validates ledger parity);
  wake branches (chatIdOf/jobIdOf, finalizeJob done, recoverStaleJobs
  re-deliver or contextRespawn); abortJob kills context jobs SILENTLY
  (nothing was promised in the lesson); turn-service trigger seam
- adapter: data-graph-context flattens to the canonical block (malformed
  parts have no model face)
- UI: the budget slider (home dock + /chat/new draft, the session-data
  strip); the quiet SystemLine "~537 · 10 propositions"; dock kind
  "Knowledge context"; model-picker PUT now spreads settings (a model
  pick can no longer clear the sibling override)

Live-smoked end to end on the real embedder + real GLM: seeded ledger
(10 propositions), session on "modal logic: possible worlds and the
necessity box" — job in the dock, queued behind the prober's question,
delivered on the answer, SystemLine in the transcript, wake turn reads
the block and continues the probe. Gates: tsc, lint, 459/459, build.

tests/record-adapter.test.ts

@@ -85,4 +85,36 @@ describe("record-adapter: system message flattening", () => {
 		]);
 		assert.equal(out.length, 0, "no model face at all — skipped");
 	});
+
+	it("the graph-context steer flattens to the canonical prior-knowledge block", () => {
+		const out = recordToPiMessages([
+			userParts([
+				{
+					type: "data-graph-context",
+					id: "ctx1",
+					data: {
+						chatId: "chat_1",
+						budget: 1024,
+						tokens: 64,
+						propositions: [
+							{ id: "n1", statement: "Water boils at 100°C.", level: "solid" },
+							{ id: "n2", statement: "Evaporation cools things.", level: null },
+						],
+					},
+				},
+			]),
+		]);
+		assert.equal(out.length, 1, "one user message — the marker part is the model face");
+		const got = texts(out);
+		assert.equal(got.length, 1);
+		assert.match(got[0]!, /Prior-knowledge context/);
+		assert.match(got[0]!, /\[solid\] Water boils at 100°C\./);
+	});
+
+	it("a malformed graph-context part has no model face — the message is skipped wholesale", () => {
+		const out = recordToPiMessages([
+			userParts([{ type: "data-graph-context", id: "ctx2", data: { chatId: "chat_1" } }]),
+		]);
+		assert.equal(out.length, 0, "parse-don't-cast: no validated data, no face");
+	});
 });