[Docs] Align B200 DeepSeek-V4-Pro balanced recipe with MegaMoE (#31373)

docs_new/src/snippets/configs/deepseek-ai/deepseek-v4.jsx

@@ -427,15 +427,16 @@ sgl-eval run aime25 \\
     {
       match: { hw: "b200", variant: "pro", quant: "fp4", strategy: "balanced", nodes: "single" },
       verified: true,
-      env: [],
+      env: [
+        "SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK=4096",
+      ],
       flags: [
         "--trust-remote-code",
         "--model-path {{MODEL_NAME}}",
         "--tp 8",
         "--dp 8",
         "--enable-dp-attention",
-        "--moe-runner-backend flashinfer_mxfp4",
-        "--disable-flashinfer-autotune",
+        "--moe-a2a-backend megamoe",
         "--chunked-prefill-size 32768",
         "--swa-full-tokens-ratio 0.1",
         "--speculative-algorithm EAGLE",
@@ -444,7 +445,6 @@ sgl-eval run aime25 \\
         "--speculative-num-draft-tokens 2",
         "--mem-fraction-static 0.92",
         "--cuda-graph-max-bs-decode 256",
-        "--deepep-config '{\"normal_dispatch\":{\"num_sms\":96},\"normal_combine\":{\"num_sms\":96}}'",
         "--host {{HOST_IP}}",
         "--port {{PORT}}",
       ],