[Cherry-pick] [0.11.0] pd proxy support ipv6 and fix proxy (#4242)

### What this PR does / why we need it?
pd proxy support ipv6, mooncake connector check whether the IPv6 address
is used and notify the user.

---------

Signed-off-by: liziyu <liziyu16@huawei.com>

examples/disaggregated_prefill_v1/load_balance_proxy_layerwise_server_example.py

@@ -88,6 +88,7 @@ import argparse
 import asyncio
 import functools
 import heapq
+import ipaddress
 import os
 import sys
 import threading
@@ -116,6 +117,12 @@ class ServerState:
         self.host = host
         self.port = port
         self.url = f'http://{host}:{port}/v1'
+        try:
+            ip = ipaddress.ip_address(self.host)
+            if isinstance(ip, ipaddress.IPv6Address):
+                self.url = f'http://[{host}]:{port}/v1'
+        except Exception:
+            pass
         self.client = httpx.AsyncClient(timeout=None,
                                         base_url=self.url,
                                         limits=httpx.Limits(
@@ -356,6 +363,9 @@ async def send_request_to_service(client: httpx.AsyncClient,
     req_data = req_data.copy()
     req_data["stream"] = False
     req_data["max_tokens"] = 1
+    req_data["min_tokens"] = 1
+    if "max_completion_tokens" in req_data:
+        req_data["max_completion_tokens"] = 1
     if "stream_options" in req_data:
         del req_data["stream_options"]
     headers = {

examples/disaggregated_prefill_v1/load_balance_proxy_server_example.py

@@ -88,6 +88,7 @@ import argparse
 import asyncio
 import functools
 import heapq
+import ipaddress
 import json
 import os
 import sys
@@ -118,6 +119,12 @@ class ServerState:
         self.host = host
         self.port = port
         self.url = f'http://{host}:{port}/v1'
+        try:
+            ip = ipaddress.ip_address(self.host)
+            if isinstance(ip, ipaddress.IPv6Address):
+                self.url = f'http://[{host}]:{port}/v1'
+        except Exception:
+            pass
         self.client = httpx.AsyncClient(timeout=None,
                                         base_url=self.url,
                                         limits=httpx.Limits(
@@ -366,6 +373,8 @@ async def send_request_to_service(client: httpx.AsyncClient,
     req_data["stream"] = False
     req_data["max_tokens"] = 1
     req_data["min_tokens"] = 1
+    if "max_completion_tokens" in req_data:
+        req_data["max_completion_tokens"] = 1
     if "stream_options" in req_data:
         del req_data["stream_options"]
     headers = {