Add diagnostics command and enhance error handling for backend states

- Introduced a new CLI command `diag` to collect diagnostics from SwarmUI/Comfy, including API, journal, and paths.
- Enhanced the `ensure_swarm_comfy_installed` function to include diagnostic script handling for errored backends.
- Updated `wait_backend_idle` to trigger diagnostics when backends are in an errored state, improving error recovery.
- Implemented fallback mechanisms for diagnostics in the `run_diagnostics` function, ensuring better visibility into backend issues.
- Added tests to validate the new diagnostic functionality and error handling, ensuring robustness in backend management.

src/gpu_rent/cli.py

@@ -712,6 +712,28 @@ def ssh() -> None:
         _die(exc)
 
 
+@app.command("diag")
+def diag_cmd(
+    lines: int = typer.Option(80, "--lines", "-n", help="Строк journal в отчёте"),
+) -> None:
+    """Снять диагностику SwarmUI/Comfy с VM (API + journal + paths)."""
+    try:
+        cfg = load_config(require_auth=True)
+        state = load_state()
+        if not state.floating_ip:
+            raise GpuRentError("нет IP — VM не поднята")
+        from gpu_rent.ready import collect_swarm_diagnostics
+
+        # lines reserved for future; swarm_diag uses fixed 80 for now
+        _ = lines
+        collect_swarm_diagnostics(cfg, state.floating_ip, console.print)
+        console.print(
+            "[dim]/mnt/swarm_data/.gpu-rent-last-diag.txt на VM[/dim]"
+        )
+    except GpuRentError as exc:
+        _die(exc)
+
+
 @app.command()
 def logs(
     unit: Optional[str] = typer.Option(