Add topic callers, town knowledge, dynamic response lengths

- 30% of callers now call about topics (prestige TV, science, poker,
  astrophotography, physics, tech, US news) instead of personal problems
- 86 curated interests weighted toward shows like Severance, Breaking Bad,
  The Wire, LOST, Westworld, etc. Removed reality TV/celebrity gossip
- 32-town knowledge base with real facts so callers don't invent landmarks
- Smart topic detection for news enrichment (keyword->search query mapping)
- Enrichment now summarizes articles naturally via LLM instead of quoting headlines
- Prompt rewrite for varied response lengths and no rehashing
- Extra weight for Animas and Lordsburg callers

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

backend/services/news.py

@@ -15,6 +15,7 @@ class NewsItem:
     title: str
     source: str
     published: str
+    content: str = ""
 
 
 class NewsService:
@@ -88,7 +89,8 @@ class NewsService:
             engines = result.get("engines", [])
             source = engines[0] if engines else ""
             published = result.get("publishedDate", "")
-            items.append(NewsItem(title=title, source=source, published=published))
+            content = result.get("content", "").strip()
+            items.append(NewsItem(title=title, source=source, published=published, content=content))
         return items
 
     def format_headlines_for_prompt(self, items: list[NewsItem]) -> str: