Enhance access link collection and backend status reporting

- Updated the `collect_access_links` function to provide clearer user-facing endpoint labels and notes, particularly for non-tunneled scenarios.
- Improved logging messages in the provisioning process to reflect the status of the SwarmUI and Ollama API, enhancing user feedback during setup.
- Added human-readable status messages for backend loading and running states, improving clarity during the waiting process.
- Updated tests to verify the new behavior and ensure accurate reporting of access links and backend statuses.

src/gpu_rent/provision.py

@@ -627,7 +627,16 @@ def provision_vm(
             _try_arm_idle_killer(cfg, host, log, conn=conn, server_id=server_id)
 
     if swarm:
-        log("SwarmUI слушает 127.0.0.1:7801 — gpu-rent tunnel")
-    if rt == "ollama":
+        from gpu_rent.access_card import print_access_card
+        from gpu_rent.term import console as rich_console
+
+        print_access_card(
+            cfg,
+            tunneled=False,
+            host=host,
+            console=rich_console,
+            title="gpu-rent · URL после tunnel (ещё ждём Idle)",
+        )
+    elif rt == "ollama":
         log(f"Ollama API → localhost:{cfg.ollama_local_port} (туннель)")
     log("Hold killer: gpu-rent hold   |  Стоп GPU: gpu-rent stop")