fix profile run for vl model (#5136)

### What this PR does / why we need it?

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: 李少鹏 <lishaopeng21@huawei.com>

tests/e2e/models/configs/Qwen3-VL-30B-A3B-Instruct.yaml

@@ -6,7 +6,6 @@ tasks:
   metrics:
   - name: "acc,none"
     value: 0.58
-max_model_len: 8192
 tensor_parallel_size: 2
 gpu_memory_utilization: 0.7
 enable_expert_parallel: True

tests/e2e/models/configs/Qwen3-VL-8B-Instruct.yaml

@@ -6,6 +6,5 @@ tasks:
   metrics:
   - name: "acc,none"
     value: 0.55
-max_model_len: 8192
 batch_size: 32
 gpu_memory_utilization: 0.7