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/paper-extract.ts

@@ -0,0 +1,16 @@
+import { Scenario } from '../types';
+
+export const scenario: Scenario = {
+  id: 'paper-extract',
+  name: 'Paper extract (optional)',
+  description: 'Extract and ingest a PDF (requires network).',
+  tools: ['paperExtract'],
+  enabled: false,
+  notes: 'Enable when network is available; can be slow.',
+  input: {
+    message: 'Add this paper: https://arxiv.org/pdf/1706.03762.pdf',
+  },
+  expect: {
+    toolsCalledSoft: ['paperExtract'],
+  },
+};