Enhance SwarmUI integration and GPU environment verification

- Updated CLI documentation to reflect the new handling of `CIVITAI_API_TOKEN`, which is now automatically passed to SwarmUI user settings during startup.
- Improved the `render_access_panel` function to include additional warnings for idle-killer failures and stack errors, enhancing user feedback.
- Introduced a new function `seed_swarmui_api_keys` to manage API key injection into SwarmUI, ensuring seamless integration with the Model Downloader.
- Enhanced GPU environment verification logic to include fail-fast checks for critical components like CUDA, improving error handling and user notifications.
- Updated tests to validate the new API key handling and access panel behavior, ensuring robustness in the integration process.

docs/cli.md

@@ -233,8 +233,12 @@ Application credential для idle-killer CLI создаёт на `up` (узко
 2. TCP 22 / SSH
 3. Backend Idle (если SwarmUI) → toast (если `NOTIFY_READY`)
 4. На VM: HTTP сервисов стека (SwarmUI `:7801` / Ollama `:11434` / llama.cpp `:8080`) — `verify_stack_on_vm`
-5. На VM: **nvidia-smi / CUDA** (+ **torch+cuda** в Comfy venv, если SwarmUI) — `verify_gpu_env`
-6. Туннель + проверка **localhost** тех же сервисов → access-card
+5. На VM: **nvidia-smi / CUDA** (fail-fast) + **torch+cuda** в Comfy venv при SwarmUI (ждём) — `verify_gpu_env`
+6. В логе: строка **`тайминг up:`** (SSH / bootstrap / Idle / verify / …)
+7. Туннель + проверка **localhost** тех же сервисов → access-card (красная рамка, если killer failed)
+
+`gpu-rent logs --unit swarm|ollama|llamacpp|killer` — фильтр journalctl.
+`gpu-rent status` — killer/hold, последний стек/GPU-env, тайминг up.
 
 Локальный порт UI: **17801** (на VM по-прежнему 7801 на loopback).