[TEST] Modify the configuration of the 310p test case (vllm-project#12007)

### What this PR does / why we need it?
Modify the configuration of the 310p test case
### Does this PR introduce _any_ user-facing change?
no
### How was this patch tested?

- vLLM version: v0.24.0
- vLLM main:
vllm-project/vllm@85c09e9

---------

Signed-off-by: zxinyue <1264904584@qq.com>
Signed-off-by: taozi01319 <68981779+taozi01319@users.noreply.github.com>

Author: taozi01319

tests/e2e/weekly/single_node/configs/Qwen3-14B-w8a8sc-310p.yaml

@@ -48,11 +48,12 @@ test_cases:
         case_type: accuracy
         dataset_path: vllm-ascend/SuperGLUE
         request_conf: vllm_api_general_chat
-        dataset_conf: BoolQ/SuperGLUE_BoolQ_gen_0_shot_cot_str
+        dataset_conf: SuperGLUE/SuperGLUE_BoolQ_gen_0_shot_cot_str
         max_out_len: 10240
         batch_size: 32
         baseline: 89.3
         threshold: 1
         temperature: 0
         top_p: 0.95
         ignore_eos: false
+

tests/e2e/weekly/single_node/configs/Qwen3-32B-w8a8sc-310p.yaml

@@ -26,7 +26,7 @@ test_cases:
       - "--load_format"
       - "sharded_state"
       - "--compilation-config"
-      - '{"cudagraph_mode":"FULL_DECODE_ONLY", "cudagraph_capture_sizes":[8,32]}'
+      - '{"cudagraph_mode":"FULL_DECODE_ONLY", "cudagraph_capture_sizes":[32]}'
       - "--additional-config"
       - '{"ascend_compilation_config":{"enable_npugraph_ex":false, "fuse_norm_quant": false}}'
       - "--default-chat-template-kwargs"
@@ -44,3 +44,14 @@ test_cases:
         request_rate: 0
         baseline: 105.26
         threshold: 0.95
+      acc:
+        case_type: accuracy
+        dataset_path: vllm-ascend/gsm8k
+        request_conf: vllm_api_general_chat
+        dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_chat_prompt
+        max_out_len: 10240
+        batch_size: 32
+        baseline: 95.07
+        threshold: 1
+        temperature: 0
+        top_p: 0.95