Add textbook formalization pipeline (extract -> LLM formalize -> elaboration check -> library assembly)

- extract_section.py: slice a book section out of a pdftotext -layout dump and
  list its labeled Theorem/Proposition/Lemma/Corollary/Definition items.
- formalize.py: LLM-formalize a section into Lean 4 candidate statements
  (preferring Mathlib vocabulary) plus a definitions manifest.
- check_statements.py: elaboration-only sanity check against a real
  Mathlib-backed Lean project (must run outside the Cowork sandbox, which has
  no Lean toolchain and can't run macOS binaries anyway).
- build_library.py: assemble solved lean-bench results into a reusable .lean
  file, reusing each winning proof verbatim.
- llm.py: standalone OpenAI-compatible chat client + .env loader, kept
  independent of lean_agent so this pipeline works without it installed.
- Source PDFs and raw/layout text dumps are gitignored (copyrighted material);
  only code and the final short Lean statement datasets are tracked.

tests/test_extract_section.py

@@ -0,0 +1,49 @@
+import unittest
+from textbook.extract_section import extract_section, labeled_statements
+
+FIXTURE = """\
+1.6. Divisibility in the Integers
+intro text
+
+Proposition 1.6.1.
+body
+
+Theorem 1.6.6. There are infinitely many prime numbers.
+Proof. blah
+36                          1. ALGEBRAIC THEMES
+
+
+Theorem 1.6.6. There are infinitely many prime numbers.
+Proof continues after a page break, repeats the label.
+
+1.7. Modular Arithmetic
+next section text
+"""
+
+
+class ExtractSectionTests(unittest.TestCase):
+    def test_extract_section_slices_between_headings(self):
+        section = extract_section(FIXTURE, '1.6. Divisibility in the Integers', '1.7. Modular Arithmetic')
+        self.assertIn('Theorem 1.6.6', section)
+        self.assertNotIn('next section text', section)
+
+    def test_extract_section_missing_heading_raises(self):
+        with self.assertRaises(ValueError):
+            extract_section(FIXTURE, '1.6. Divisibility in the Integers', '9.9. Does Not Exist')
+
+    def test_labeled_statements_dedupes_and_orders(self):
+        section = extract_section(FIXTURE, '1.6. Divisibility in the Integers', '1.7. Modular Arithmetic')
+        items = labeled_statements(section, '1.6.')
+        labels = [item['label'] for item in items]
+        # Theorem 1.6.6 appears twice (page-break repeat) but must be kept once, and
+        # book order (1.6.1 before 1.6.6) must be preserved.
+        self.assertEqual(labels, ['1.6.1', '1.6.6'])
+
+    def test_labeled_statements_filters_by_prefix(self):
+        section = extract_section(FIXTURE, '1.6. Divisibility in the Integers', '1.7. Modular Arithmetic')
+        items = labeled_statements(section, '1.7.')
+        self.assertEqual(items, [])
+
+
+if __name__ == '__main__':
+    unittest.main()