Support incremental streaming of logprob/token_ids between scheduler and detokenizer (#6225)

Co-authored-by: SangBin Cho <rkooo567@gmail.com>

python/sglang/srt/server_args.py

@@ -98,6 +98,7 @@ class ServerArgs:
     show_time_cost: bool = False
     enable_metrics: bool = False
     decode_log_interval: int = 40
+    enable_request_time_stats_logging: bool = False
 
     # API related
     api_key: Optional[str] = None
@@ -785,6 +786,12 @@ class ServerArgs:
             default=ServerArgs.decode_log_interval,
             help="The log interval of decode batch.",
         )
+        parser.add_argument(
+            "--enable-request-time-stats-logging",
+            action="store_true",
+            default=ServerArgs.enable_request_time_stats_logging,
+            help="Enable per request time stats logging",
+        )
 
         # API related
         parser.add_argument(