[V0 Deprecation] Remove VLLM_USE_V1 from docs and scripts (#26336)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

examples/online_serving/disaggregated_serving_p2p_nccl_xpyd/disagg_example_p2p_nccl_xpyd.sh

@@ -166,7 +166,7 @@ main() {
         local kv_port=$((21001 + i))
 
         echo "  Prefill server $((i+1)): GPU $gpu_id, Port $port, KV Port $kv_port"
-        CUDA_VISIBLE_DEVICES=$gpu_id VLLM_USE_V1=1 vllm serve $MODEL \
+        CUDA_VISIBLE_DEVICES=$gpu_id vllm serve $MODEL \
         --enforce-eager \
         --host 0.0.0.0 \
         --port $port \
@@ -194,7 +194,7 @@ main() {
         local kv_port=$((22001 + i))
 
         echo "  Decode server $((i+1)): GPU $gpu_id, Port $port, KV Port $kv_port"
-        VLLM_USE_V1=1 CUDA_VISIBLE_DEVICES=$gpu_id vllm serve $MODEL \
+        CUDA_VISIBLE_DEVICES=$gpu_id vllm serve $MODEL \
         --enforce-eager \
         --host 0.0.0.0 \
         --port $port \

examples/online_serving/elastic_ep/serve_deepseek_v2.sh

@@ -55,7 +55,6 @@ done
 echo "Starting vLLM server for $MODEL_NAME with data parallel size: $DATA_PARALLEL_SIZE and redundant experts: $REDUNDANT_EXPERTS"
 
 export RAY_DEDUP_LOGS=0
-export VLLM_USE_V1=1
 export VLLM_ALL2ALL_BACKEND="pplx"
 export VLLM_USE_DEEP_GEMM=1
 

examples/online_serving/openai_chat_completion_client_with_tools_required.py

@@ -5,7 +5,7 @@ To run this example, you can start the vLLM server
 without any specific flags:
 
 ```bash
-VLLM_USE_V1=0 vllm serve unsloth/Llama-3.2-1B-Instruct \
+vllm serve unsloth/Llama-3.2-1B-Instruct \
     --structured-outputs-config.backend outlines
 ```
 

examples/online_serving/ray_serve_deepseek.py

@@ -36,7 +36,6 @@ llm_config = LLMConfig(
     },
     # Set to the node's accelerator type.
     accelerator_type="H100",
-    runtime_env={"env_vars": {"VLLM_USE_V1": "1"}},
     # Customize engine arguments as required (for example, vLLM engine kwargs).
     engine_kwargs={
         "tensor_parallel_size": 8,