[Cherry Pick from pr#3981][0.11.0][P/D]Make kv-transfer env variable take effect & Fix load-balance proxy (#3983)

### What this PR does / why we need it?
Make kv-transfer env variable take effect & Fix load-balance proxy.
Cherry Pick from #3981

---------
Signed-off-by: nwpu-zxr <zhouxuerong2@huawei.com>

vllm_ascend/distributed/mooncake_connector.py

@@ -2,6 +2,7 @@
 import contextlib
 import hashlib
 import math
+import os
 import queue
 import random
 import struct
@@ -33,6 +34,7 @@ from vllm.v1.request import RequestStatus
 import vllm_ascend.envs as envs_ascend
 from vllm_ascend.ascend_config import get_ascend_config, init_ascend_config
 from vllm_ascend.distributed.mooncake.transfer_engine import get_global_te
+from vllm_ascend.distributed.utils import get_transfer_timeout_value
 
 if TYPE_CHECKING:
     from vllm.attention.backends.abstract import AttentionMetadata
@@ -849,6 +851,8 @@ class MooncakeConnectorWorker:
 
     def __init__(self, vllm_config: VllmConfig, engine_id: str):
         self._get_prefill_decode_size(vllm_config)
+        os.environ["ASCEND_TRANSFER_TIMEOUT"] = str(
+            get_transfer_timeout_value())
         if self._prefill_tp_size < self._decode_tp_size:
             raise ValueError(
                 f"prefill_tp_size: {self._prefill_tp_size} must be greater than"