Add multilingual audit CI pipeline + extract mandelblog_content_guard

mandelblog_content_guard/normalizers/nl.py

@@ -0,0 +1,15 @@
+from __future__ import annotations
+
+import re
+
+
+NL_PHRASE_REPLACEMENTS = {
+    "PLAN": "PLAN",
+}
+
+
+def normalize_nl_text(text: str, field_path: str = "") -> str:
+    cleaned = text
+    for source, target in NL_PHRASE_REPLACEMENTS.items():
+        cleaned = cleaned.replace(source, target)
+    return re.sub(r"\s+", " ", cleaned).strip()