[CI] Optimize nightly CI (#3898)

### What this PR does / why we need it?
This patch mainly fix the the problem of not being able to determine the
exit status of the pod's entrypoint script and some other tiny
optimizations:
1. Shorten wait for server timeout
2. fix typo
3. fix the issue of ais_bench failing to correctly access the proxy URL
in a PD separation scenario.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?


- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: wangli <wangli858794774@gmail.com>

tests/e2e/nightly/multi_node/test_multi_node.py

@@ -118,6 +118,11 @@ async def test_multi_node() -> None:
                 port = proxy_port if disaggregated_prefill else server_port
                 # aisbench test
                 aisbench_cases = [acc_cmd, perf_cmd]
-                run_aisbench_cases(local_model_path, port, aisbench_cases)
+                run_aisbench_cases(local_model_path,
+                                   port,
+                                   aisbench_cases,
+                                   host_ip=config.cluster_ips[0])
             else:
-                remote_server.hang_until_terminated()
+                # for the nodes except master, should hang until the task complete
+                master_url = f"http://{config.cluster_ips[0]}:{server_port}/health"
+                remote_server.hang_until_terminated(master_url)