chore: update qwen-agentworld-35b recipe with tool call args and model name

- Remove required_nodes and distributed_backend (single-node default)
- Add --enable-auto-tool-choice, --tool-call-parser qwen3_xml
- Add --served-model-name qwen-agentworld-35b

recipes/qwen-agentworld-35b/recipe.toml

@@ -2,10 +2,8 @@ engine = "vllm"
 name = "Qwen-AgentWorld-35B-A3B (vLLM)"
 docker_image = "ghcr.io/timothystewart6/vllm-gb10:v0.23.0-gb10.1"
 model = "Qwen/Qwen-AgentWorld-35B-A3B"
-required_nodes = 1
 tensor_parallel_size = 1
 pipeline_parallel_size = 1
-distributed_backend = "mp"
 
 # Solo: ~70 GB BF16 weights, only 10 GA layers need KV cache (DeltaNet is linear).
 # Use: sm121-launcher serve --recipe qwen-agentworld-35b --node dgx1.lan
@@ -15,11 +13,13 @@ distributed_backend = "mp"
 entrypoint = ["vllm", "serve"]
 
 args = """
---port 30000
 --max-model-len 262144
 --reasoning-parser qwen3
 --language-model-only
 --trust-remote-code
 --gpu-memory-utilization 0.90
 --enable-prefix-caching
+--enable-auto-tool-choice
+--tool-call-parser qwen3_xml
+--served-model-name qwen-agentworld-35b
 """