vLLM AsyncEngineArgs defaults for the ops-relevant engine and scheduler knobs

Defaults transcribed from the dataclass fields: shutdown_timeout: int = 0; max_num_queued_reqs: int | None = None; max_num_queued_tokens: int | None = None; max_num_seqs: int | None = None; max_num_batched_tokens: int | None = None; max_num_scheduled_tokens: int | None = None; max_num_active_seqs inherited from SchedulerConfig; long_prefill_token_threshold and long_prefill_token_threshold_adaptive inherited from SchedulerConfig; gpu_memory_utilization and kv_cache_memory_bytes inherited from CacheConfig. The load-bearing fact for an ops record is that shutdown_timeout, max_num_queued_reqs and max_num_queued_tokens all default to off (0 or None), so a stock deployment has no in-flight drain window and no queue admission bound.