fix: 里程碑2 最终评审 must-fix（header_row 接线 + formatting 回填）

tests/test_excel_parser.py

@@ -24,3 +24,36 @@ def test_parse_free_text_sheet(tmp_path):
     result = ExcelParser().parse(path)
     assert len(result.tables) == 1
     assert result.tables[0].extraction_method == "llm_from_free_text"
+
+
+def test_parse_uses_detected_header_row(tmp_path):
+    # 标题行（单格）在首行，真实表头在第二行
+    wb = new_workbook({
+        "機能一覧": [["機能一覧"], ["機能ID", "機能名"], ["A001", "社員登録"]],
+    })
+    path = save_workbook(tmp_path, wb)
+    result = ExcelParser().parse(path)
+    t = result.tables[0]
+    assert t.headers == ["機能ID", "機能名"]
+    assert len(t.rows) == 1
+    assert t.rows[0]["機能ID"].value == "A001"
+
+
+def test_parse_attaches_strikethrough_formatting(tmp_path):
+    from copy import copy
+    from openpyxl import Workbook
+    wb = Workbook()
+    ws = wb.active
+    ws.title = "機能一覧"
+    ws["A1"] = "機能ID"
+    ws["B1"] = "機能名"
+    ws["A2"] = "A001"
+    ws["B2"] = "社員登録"
+    font = copy(ws["A2"].font)
+    font.strike = True
+    ws["A2"].font = font
+    path = save_workbook(tmp_path, wb)
+    result = ExcelParser().parse(path)
+    cv = result.tables[0].rows[0]["機能ID"]
+    assert cv.formatting is not None
+    assert cv.formatting.strikethrough is True