fix: improve gemini compatibility and cache flow

1. add Gemini /models and /v1beta/models discovery endpoints for newapi compatibility

2. persist tmp cache in Docker and lock cache_timeout=0 behavior with regression tests

3. refine image generation status flow to show captcha verification after uploads

docker-compose.proxy.yml

@@ -8,6 +8,7 @@ services:
       - "38000:8000"
     volumes:
       - ./data:/app/data
+      - ./tmp:/app/tmp
       - ./config/setting_warp.toml:/app/config/setting.toml
     environment:
       - PYTHONUNBUFFERED=1