Refactor codebase following Clean Code principles and add 229 tests

- Extract helpers to reduce function sizes (listing_tasks, app.py, query.py, listing_fetcher)
  - Replace nonlocal mutations with _PipelineState dataclass in listing_tasks
  - Fix bugs: isinstance→equality check in repository, verify_exp for OIDC tokens
  - Consolidate duplicate filter methods in listing_repository
  - Move hardcoded config to env vars with backward-compatible defaults
  - Simplify CLI decorator to auto-build QueryParameters
  - Add deprecation docstring to data_access.py
  - Test count: 158 → 387 (all passing)

crawler/tests/unit/test_floorplan.py

@@ -0,0 +1,104 @@
+"""Unit tests for rec/floorplan.py."""
+from unittest.mock import patch
+import numpy as np
+from PIL import Image
+import pytest
+
+from rec.floorplan import extract_total_sqm, improve_img_for_ocr, calculate_ocr
+
+
+class TestExtractTotalSqm:
+
+    def test_normal_value(self) -> None:
+        assert extract_total_sqm("Total area: 75.5 sq m") == 75.5
+
+    def test_multiple_values_returns_max_in_range(self) -> None:
+        assert extract_total_sqm("Room 1: 20 sqm, Total: 65 sq m") == 65.0
+
+    def test_no_match_returns_none(self) -> None:
+        assert extract_total_sqm("No area info") is None
+
+    def test_below_minimum_returns_none(self) -> None:
+        assert extract_total_sqm("Area: 15 sq m") is None
+
+    def test_above_maximum_returns_none(self) -> None:
+        assert extract_total_sqm("Area: 200 sq m") is None
+
+    def test_edge_just_above_min(self) -> None:
+        assert extract_total_sqm("Area: 30.1 sq m") == 30.1
+
+    def test_edge_just_below_max(self) -> None:
+        assert extract_total_sqm("Area: 159.9 sq m") == 159.9
+
+    def test_exactly_at_min_boundary_returns_none(self) -> None:
+        # MIN_SQM < sqm, so 30 is not strictly greater than 30
+        assert extract_total_sqm("Area: 30 sq m") is None
+
+    def test_exactly_at_max_boundary_returns_none(self) -> None:
+        # sqm < MAX_SQM, so 160 is not strictly less than 160
+        assert extract_total_sqm("Area: 160 sq m") is None
+
+    def test_format_sq_dot_m(self) -> None:
+        assert extract_total_sqm("Area: 80 sq. m") == 80.0
+
+    def test_format_sqm_no_space(self) -> None:
+        assert extract_total_sqm("Area: 80sqm") == 80.0
+
+    def test_format_sq_m_with_space(self) -> None:
+        assert extract_total_sqm("Area: 80 sq m") == 80.0
+
+    def test_empty_string(self) -> None:
+        assert extract_total_sqm("") is None
+
+    def test_multiple_valid_values_returns_max(self) -> None:
+        assert extract_total_sqm("Living: 40 sq m, Total: 100 sq m") == 100.0
+
+
+class TestImproveImgForOcr:
+
+    def test_produces_valid_pil_image(self) -> None:
+        # Create a small test image (50x50 white image)
+        img = Image.fromarray(np.ones((50, 50, 3), dtype=np.uint8) * 200)
+        result = improve_img_for_ocr(img)
+        assert isinstance(result, Image.Image)
+        # Result should be a grayscale (thresholded) image
+        assert result.mode == "L"
+
+    def test_output_dimensions_scaled(self) -> None:
+        img = Image.fromarray(np.ones((100, 100, 3), dtype=np.uint8) * 128)
+        result = improve_img_for_ocr(img)
+        # After 1.2x resize, 100 -> 120
+        assert result.size[0] == 120
+        assert result.size[1] == 120
+
+
+class TestCalculateOcr:
+
+    def test_invalid_path_raises_file_not_found(self) -> None:
+        with pytest.raises(FileNotFoundError):
+            calculate_ocr("/nonexistent/path/to/image.png")
+
+    def test_returns_sqm_from_first_pass(self, tmp_path) -> None:  # type: ignore[no-untyped-def]
+        # Create a real image file so the path check passes
+        image_file = tmp_path / "test.png"
+        Image.fromarray(np.ones((10, 10, 3), dtype=np.uint8)).save(str(image_file))
+
+        with patch("pytesseract.image_to_string", return_value="Total: 85 sq m"):
+            result_sqm, result_text = calculate_ocr(str(image_file))
+
+        assert result_sqm == 85.0
+        assert result_text == "Total: 85 sq m"
+
+    def test_falls_back_to_improved_image(self, tmp_path) -> None:  # type: ignore[no-untyped-def]
+        image_file = tmp_path / "test.png"
+        Image.fromarray(np.ones((10, 10, 3), dtype=np.uint8)).save(str(image_file))
+
+        # First call returns no sqm data, second (on improved image) returns valid data
+        with patch("pytesseract.image_to_string", side_effect=[
+            "No area info here",
+            "Total: 72 sq m",
+        ]):
+            result_sqm, result_text = calculate_ocr(str(image_file))
+
+        assert result_sqm == 72.0
+        assert result_text == "Total: 72 sq m"