fix: seed AI model config at runtime

packaging/default_config/ai_models.json

@@ -4,7 +4,7 @@
       "name": "GPT Image 2",
       "url": "https://api.vectorengine.ai/v1/images/edits",
       "model": "gpt-image-2",
-      "api_key": "sk-JgzW0CWem3ybS6JtmiML2s9IbDD68hxVTrPXnpZnQRbxpC9b",
+      "api_key": "",
       "api_type": "images_edits",
       "timeout_seconds": 0,
       "connect_timeout_seconds": 30,
@@ -14,7 +14,7 @@
       "name": "Nano Banana 2",
       "url": "https://api.vectorengine.ai/v1/chat/completions",
       "model": "gemini-3.1-flash-image-preview",
-      "api_key": "sk-7x0SvcyoGsVpWPW9V622y6vsiG6BvaNJLCGKKUj9StnkZbwI",
+      "api_key": "",
       "api_type": "auto",
       "timeout_seconds": 0,
       "connect_timeout_seconds": 30,