BYOK provider embeddings (wave 6 chunk B): kill local transformers

Replace the @huggingface/transformers local embedder (~316MB ONNX cache)
with an EmbeddingPort seam: graph.embeddingModel (nullable "provider:id"
slug) selects an OpenRouter/OpenAI/Gemini adapter, resolved fresh per
call from settings + the auth store so key rotations land immediately.

- src/server/embeddings.ts rewritten: slug grammar, injected-fetch
  adapters (OpenRouter input_type, Gemini taskType batching, OpenAI
  index-ordered rows), honest resolution (null = skip-with-info, set
  model with missing key = honest error), embedNewNodes with model-change
  re-embed wipe + ghost pruning + usage steps, federated catalog
  (OpenRouter /api/v1/embeddings/models — VERIFIED live, the inverse
  path 404s; OpenAI text-embedding- filter; Gemini embedContent filter).
- GET /api/models/embeddings: federated catalog, empty when no keys.
- Extraction tail records stages.embedding usage into the run's money
  channel; EMBEDDING_PRICES overlay in core/prices.ts (outside the
  sync-generated block), callCost falls back to it; cost-summary folds
  the fourth stage; graph-context injection resolves its embedder per
  run, fails honestly when unconfigured, records query tokens.
- Settings: GraphSettings.embeddingModel (null factory), section parser
  validates slug shape (id may contain colons — OpenRouter :free
  suffixes, live catalog fact).
- Removal: @huggingface/transformers uninstalled, PATHS.modelCache and
  onnx tracing entries deleted, @huggingface tracing-excludes guard
  added, stale sharp/libvips comments updated to historical.

Live-verified on port 3004: catalog federates 36 rows (33 OpenRouter +
3 OpenAI), embedNewNodes embedded 10 nodes via the free liquid/lfm-2.5
model (dim 1024, 435 prompt tokens, 1 call, $0), settings PUT rejects
bad slugs with 400.

Gates: typecheck, lint, 895/895 tests, build.

tests/cost-summary.test.ts

@@ -96,6 +96,39 @@ describe("graphCostSummary (the cost pill's fold)", () => {
 		assert.equal(s.unpricedRuns, 1); // run-level: only the both-stages run
 	});
 
+	test("the embedding stage (wave 6): folds when present, absent when the run embedded nothing", () => {
+		const EMBED_MODEL = "openai:text-embedding-3-small";
+		const embStep = callCost(step(5000, 0), EMBED_MODEL)!;
+		assert.ok(embStep > 0, "the overlay prices the known embedding models");
+		const withEmbed = extractionRun(100, PRICED, PRICED);
+		withEmbed.stages.embedding = { model: EMBED_MODEL, steps: [step(5000, 0)] };
+		const s = graphCostSummary({
+			chats: [{ title: "a", extractions: [withEmbed] }, { title: "b", extractions: [extractionRun(200, PRICED, PRICED)] }],
+			sources: [],
+			naming: [],
+		});
+		assert.equal(s.stages.embedding.runs, 1, "only the run that embedded");
+		assert.equal(s.stages.embedding.dollars, embStep);
+		assert.equal(s.stages.embedding.tokens.inputTokens, 5000);
+		assert.equal(s.stages.extraction.runs, 2, "the two-stage era still folds");
+		assert.equal(s.stages.extraction.runs, 2);
+		// Run-level: the embedding dollars ride the run's total.
+		const extStep = callCost(step(1000, 200), PRICED)!;
+		const audStep = callCost(step(2000, 100), PRICED)!;
+		assert.equal(s.runs.find((r) => r.startedAt === 100)!.dollars, extStep + audStep + embStep);
+		assert.equal(s.runs.find((r) => r.startedAt === 200)!.dollars, extStep + audStep);
+		assert.equal(s.totalDollars, (extStep + audStep) * 2 + embStep);
+	});
+
+	test("an UNPRICED embedding model stays honest: counted, never a silent zero; a priced sibling saves the run total", () => {
+		const unpricedEmbed = extractionRun(100, PRICED, "unknown-model");
+		unpricedEmbed.stages.embedding = { model: "some/new-embed", steps: [step(900, 0)] };
+		const s = graphCostSummary({ chats: [{ title: "x", extractions: [unpricedEmbed] }], sources: [], naming: [] });
+		assert.equal(s.stages.embedding.unpricedRuns, 1);
+		assert.equal(s.stages.embedding.dollars, 0);
+		assert.notEqual(s.runs[0]!.dollars, null, "the priced extractor saves the run total");
+	});
+
 	test("recent runs cap: newest survive", () => {
 		const extractions = [1, 2, 3, 4].map((n) => extractionRun(n * 100, PRICED, PRICED));
 		const s = graphCostSummary({