feat(db): add SQLite database persistence for insights

Implement local database storage to automatically save all sent insights
with metadata including topic, content, AI provider, model, and timestamp.
Add database module with initialization and query functions, integrate
with main job flow, and update documentation with database schema details.

main.py

@@ -6,6 +6,7 @@ import openai
 import schedule
 import time
 from datetime import datetime
+from database import init_db, save_insight
 
 load_dotenv()
 
@@ -45,6 +46,10 @@ elif OPENAI_API_KEY:
 else:
     raise ValueError("Nenhuma chave de IA configurada no .env")
 
+init_db()
+
+CURRENT_PROVIDER = DEFAULT_AI_PROVIDER
+
 TOPICS = [t.strip() for t in TOPICS_RAW.split(",") if t.strip()]
 if not TOPICS:
     raise ValueError("Nenhum tópico informado no .env (variável TOPICS)")
@@ -148,6 +153,11 @@ def job():
     print(f"[{now.strftime('%Y-%m-%d %H:%M:%S')}] Tópico escolhido: {topic}")
     print(f"[{now.strftime('%Y-%m-%d %H:%M:%S')}] Gerando insight...")
     insight = generate_insight(topic)
+    try:
+        save_insight(topic=topic, content=insight, provider=CURRENT_PROVIDER, model=DEFAULT_AI_MODEL)
+        print(f"[{now.strftime('%Y-%m-%d %H:%M:%S')}] Insight salvo no banco de dados.")
+    except Exception as e:
+        print(f"[{now.strftime('%Y-%m-%d %H:%M:%S')}] Erro ao salvar no banco: {e}")
     send_telegram_message(insight, topic)