add deep search benchmark

deepsearch_result.jsonl

@@ -7,6 +7,7 @@
 {"org": "Alibaba","link": "https://github.com/Alibaba-NLP/WebAgent","method": "WebShaper", "model": "qwq-32b", "overall": 0.497}
 {"org": "Alibaba","link": "https://github.com/Alibaba-NLP/WebAgent","method": "WebShaper", "model": "qwen2.5-72b-instruct", "overall": 0.522}
 {"org": "Tencent","link": "https://github.com/TencentCloudADP/youtu-agent","method": "Youtu-agent", "model": "deepseek-v3.1", "overall": 0.7147}
+{"org": "Tencent","link": "https://github.com/TencentCloudADP/youtu-agent","method": "Youtu-agent", "model": "deepseek-v3-0324", "overall": 0.6071}
 {"org": "Miromind","link": "https://github.com/MiroMindAI/MiroThinker","method": "MiroThinker-SFT-v0.1", "model": "qwen3-8b", "overall": 0.413}
 {"org": "Miromind","link": "https://github.com/MiroMindAI/MiroThinker","method": "MiroThinker-DPO-v0.1", "model": "qwen3-8b", "overall": 0.457}
 {"org": "Miromind","link": "https://github.com/MiroMindAI/MiroThinker","method": "MiroThinker-SFT-v0.1", "model": "qwen3-32b", "overall": 0.457}