fix: Fix salary Excel column detection for index headers (#64)

改动点：
修复 [tools/convert_salary_excel.py (line 47)](/Users/alwin/ai-job-search/tools/convert_salary_excel.py:47) 里列类型识别的问题：之前 n 被当作任意子串匹配，导致 Index / Engineering Index 这类列会被误判成 count。
同步修复类别名生成，避免 Engineering Count 里的 n 被删坏。
把 openpyxl 缺失报错延迟到实际运行转换命令时，这样纯函数可以被单元测试导入。
新增 [tests/test_convert_salary_excel.py (line 25)](/Users/alwin/ai-job-search/tests/test_convert_salary_excel.py:25)，覆盖 index/count 识别和 worksheet 解析。

Co-authored-by: Alwin.Zhang <alwin.zhang420@gmail.com>

tests/__init__.py

@@ -0,0 +1 @@
+

tests/test_convert_salary_excel.py

@@ -0,0 +1,52 @@
+import unittest
+from types import SimpleNamespace
+
+from tools.convert_salary_excel import detect_column_type, parse_sheet
+
+
+class FakeWorksheet:
+    title = "Sheet1"
+
+    def __init__(self, rows):
+        self.rows = rows
+
+    def iter_rows(self, min_row=1, max_row=None, values_only=False):
+        rows = self.rows[min_row - 1:max_row]
+        for row in rows:
+            if values_only:
+                yield row
+            else:
+                yield [SimpleNamespace(value=value) for value in row]
+
+    def __getitem__(self, row_number):
+        return [SimpleNamespace(value=value) for value in self.rows[row_number - 1]]
+
+
+class DetectColumnTypeTests(unittest.TestCase):
+    def test_index_headers_are_not_misclassified_as_count(self):
+        for header in ("Index", "Salary Index", "Engineering Index", "Median salary"):
+            with self.subTest(header=header):
+                self.assertEqual(detect_column_type(header), "index")
+
+    def test_single_letter_n_only_matches_as_a_token(self):
+        self.assertEqual(detect_column_type("Employee n"), "count")
+        self.assertEqual(detect_column_type("Engineering"), None)
+
+    def test_count_headers_still_match_common_labels(self):
+        for header in ("Count", "Engineering Count", "Antal medarbejdere"):
+            with self.subTest(header=header):
+                self.assertEqual(detect_column_type(header), "count")
+
+    def test_parse_sheet_preserves_category_name_with_letter_n(self):
+        ws = FakeWorksheet([
+            ("Company", "Engineering Count", "Engineering Index"),
+            ("Example Corp", 12, 105.5),
+        ])
+
+        companies = parse_sheet(ws)
+
+        self.assertEqual(companies[0]["categories"]["engineering"], {"count": 12, "index": 105.5})
+
+
+if __name__ == "__main__":
+    unittest.main()