[Misc] Update pooling example (#5002)

### What this PR does / why we need it?
Since the param `task` has been depprecated, we should use the latest
unified standard parameters for pooling models, this should be more
clear


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

---------

Signed-off-by: wangli <wangli858794774@gmail.com>

examples/offline_embed.py

@@ -44,7 +44,7 @@ def main():
     ]
     input_texts = queries + documents
 
-    model = LLM(model="Qwen/Qwen3-Embedding-0.6B", task="embed")
+    model = LLM(model="Qwen/Qwen3-Embedding-0.6B", runner="pooling")
 
     outputs = model.embed(input_texts)
     embeddings = torch.tensor([o.outputs.embedding for o in outputs])