Move PoolerConfig from config/__init__.py to config/pooler.py (#25181)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

examples/online_serving/openai_embedding_long_text/README.md

@@ -42,7 +42,7 @@ python client.py
 
 ### Server Configuration
 
-The key parameters for chunked processing are in the `--override-pooler-config`:
+The key parameters for chunked processing are in the `--pooler-config`:
 
 ```json
 {

examples/online_serving/openai_embedding_long_text/client.py

@@ -13,7 +13,7 @@ Prerequisites:
    
    # MEAN pooling (processes all chunks, recommended for complete coverage)
    vllm serve intfloat/multilingual-e5-large \
-     --override-pooler-config \
+     --pooler-config \
       '{"pooling_type": "MEAN", "normalize": true, ' \
       '"enable_chunked_processing": true, "max_embed_len": 3072000}' \
      --served-model-name multilingual-e5-large \
@@ -23,7 +23,7 @@ Prerequisites:
 
    # OR CLS pooling (native CLS within chunks, MEAN aggregation across chunks)
    vllm serve BAAI/bge-large-en-v1.5 \
-     --override-pooler-config \
+     --pooler-config \
       '{"pooling_type": "CLS", "normalize": true, ' \
       '"enable_chunked_processing": true, "max_embed_len": 1048576}' \
      --served-model-name bge-large-en-v1.5 \

examples/online_serving/openai_embedding_long_text/service.sh

@@ -103,7 +103,7 @@ POOLER_CONFIG="{\"pooling_type\": \"$POOLING_TYPE\", \"normalize\": true, \"enab
 vllm serve "$MODEL_NAME" \
   --tensor-parallel-size "$GPU_COUNT" \
   --enforce-eager \
-  --override-pooler-config "$POOLER_CONFIG" \
+  --pooler-config "$POOLER_CONFIG" \
   --served-model-name ${MODEL_CODE} \
   --api-key "$API_KEY" \
   --trust-remote-code \