fix(910c/glm52): 镜像与 client 修正 - 部署在 910c.2 的 GLM5.2-tuned 镜像

- profile DOCKER_IMAGE 改回 local/vllm-ascend:0.23-a3-20260718-sglang
  (910c.1 的 glm5.2-a3-openeuler 缺 expert_map_manager 模块无法启动)
- config.env DOCKER_CLIENT_IMAGE 指向 910c.2 本地 tuned 镜像(带 bench_serving)
- 实测: 910c.2 TP8/DP2 smoke 40/40, TTFT 1062ms, TPOT 51ms

deploy/profiles/910c/glm52_910c_vllm_tp_dp_matrix.env

@@ -9,7 +9,7 @@ EXPERIMENT=glm52_910c_vllm_tp_dp_matrix
 MODEL_NAME=GLM-5.2
 ENGINE=vllm
 RUNTIME=docker
-DOCKER_IMAGE=quay.io/ascend/vllm-ascend:glm5.2-a3-openeuler
+DOCKER_IMAGE=local/vllm-ascend:0.23-a3-20260718-sglang
 CONTAINER_NAME=vllm-ascend-glm52-910c_tp${TP}_dp${DP}
 MODEL_PATH=/mnt/models/GLM-5.2-w4a8c8
 SERVED_MODEL_NAME=glm-5.2

experiments/910c/glm52_910c_vllm_tp_dp_matrix/config.env

@@ -33,7 +33,7 @@ DOCKER_IMAGE="${DOCKER_IMAGE:-local/vllm-ascend:0.23-a3-20260718-sglang}"
 # (local/vllm-ascend:0.23-a3-*-sglang, see envs/ASCEND_910C_ENV_SETUP.md §8.1)
 # carry sglang 0.5.2 bench_serving, so the client runs as a standalone container
 # (TORCH_DEVICE_BACKEND_AUTOLOAD=0 injected by the runner, no NPU needed).
-DOCKER_CLIENT_IMAGE="${DOCKER_CLIENT_IMAGE:-local/vllm-ascend:0.23-a3-dsv4-sglang}"
+DOCKER_CLIENT_IMAGE="${DOCKER_CLIENT_IMAGE:-local/vllm-ascend:0.23-a3-20260718-sglang}"
 USE_DOCKER_CLIENT="${USE_DOCKER_CLIENT:-1}"
 
 # Device selection. ASCEND_VISIBLE_DEVICES selects NPU cards 0..7; the Ascend