Filter out garbage API data (language codes, metadata)

- Reject authors with <= 3 chars (language codes like "es", "it", "no")
- Reject known non-review authors ("google", "maps", etc.)
- Reject timestamps that are URLs or very short strings

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

modules/scraper_clean.py

@@ -25,6 +25,15 @@ def parse_api_review(raw: list) -> dict:
         if not (1 <= rating <= 5):
             return None
 
+        # Filter out garbage data (language codes, metadata, etc.)
+        if len(author) <= 3:  # Real names are longer than 3 chars
+            return None
+        if author.lower() in ['google', 'maps', 'reviews', 'es', 'en', 'it', 'no', 'de', 'fr', 'pt']:
+            return None
+        # Timestamp should look like a date, not a URL or language code
+        if timestamp and ('http' in str(timestamp) or len(str(timestamp)) <= 3):
+            return None
+
         # Owner response
         owner_response = None
         for idx in [9, 18]: