feat: multi-domain failover for provider accounts

Each ProviderAccount can now have multiple base URLs (provider_urls table).
On stream failure, BroadcastGroup cycles to the next domain immediately
with no wait; backs off only after all domains have been tried once.
Background health checker pings every domain every 5 min via player_api.php
and updates status/response_ms. Admin UI shows domain list with color-coded
status badges and a "Verificar todos" button per provider.

backend/database.py

@@ -1,5 +1,6 @@
 from sqlalchemy.ext.asyncio import create_async_engine, AsyncSession, async_sessionmaker
 from sqlalchemy.orm import DeclarativeBase
+from sqlalchemy import text
 from .config import settings
 
 engine = create_async_engine(
@@ -27,5 +28,15 @@ async def get_db():
 
 async def init_db():
     from .models import provider, channel, user, epg, custom_category, log, notification, jellyfin  # noqa: ensure all models imported
+    from .models.provider import ProviderUrl  # noqa: ensure ProviderUrl is registered
     async with engine.begin() as conn:
         await conn.run_sync(Base.metadata.create_all)
+        # Migrate: create provider_urls entries from existing base_url if not already done
+        await conn.execute(text("""
+            INSERT INTO provider_urls (provider_account_id, url, priority, is_active, status)
+            SELECT id, base_url, 0, true, 'unknown'
+            FROM provider_accounts
+            WHERE id NOT IN (SELECT DISTINCT provider_account_id FROM provider_urls WHERE provider_account_id IS NOT NULL)
+            ON CONFLICT DO NOTHING
+        """))
+        await conn.commit()