[NV] llm-d-vllm: optimize DSv4-Pro GB200 recipe configs

[ilmarkov]

Tune decode and prefill vLLM flags based on validated GB200 NVL72 benchmarks showing +2-43% tok/s/GPU and 15-25% lower TPOT:

Decode (mid-curve-megamoe):

• gpu-memory-utilization 0.85 -> 0.9 (larger KV cache pool)
• Add max-model-len 9280 (tight ISL8192+OSL1024 bound)
• max-num-seqs/batched-tokens/cudagraph-capture 512 -> 1024
• Disable NCCL symmetric memory (standard NVLink path faster)
• Add no-enable-flashinfer-autotune, rust frontend

Prefill (both recipes):

• gpu-memory-utilization 0.9 -> 0.95
• Add max-model-len 9280, max-num-seqs 16, max-num-batched-tokens 32768
• Disable NCCL symmetric memory, enable rust frontend
• Enable randomize-dp-dummy-inputs

EPP (both recipes):

• Switch from max-score-picker to weighted-random-picker (threshold=0.1) for better load distribution under high concurrency

Low-latency decode: disable NCCL symmetric memory, add rust frontend.