feat: sync runtime search and schema quality updates from app repo

- port retrieval, validation, and eval improvements relevant to os
- align prompts and dimensions with the flat single-agent model
- replace the old eval suite with the focused core scenarios

Generated with Codex

tests/evals/scenarios/archive/create-edge.ts

@@ -0,0 +1,15 @@
+import { Scenario } from '../types';
+
+export const scenario: Scenario = {
+  id: 'create-edge',
+  name: 'Create edge between two new nodes',
+  description: 'Create two nodes and connect them with an edge.',
+  tools: ['createNode', 'createEdge'],
+  input: {
+    message: 'Create nodes titled "Eval Edge A" and "Eval Edge B" (one sentence each), then create an edge from A to B labeled "related".',
+  },
+  expect: {
+    toolsCalledSoft: ['createNode', 'createEdge'],
+    responseContainsSoft: ['edge', 'connected'],
+  },
+};