stage19: performance optimierung — LLM concurrency semaphore, priority (HIGH/NORMAL/LOW), limiter

src/nsct/api/rest_research.py

@@ -14,6 +14,7 @@ from pydantic import BaseModel, Field
 from nsct.logging_config import set_request_ctx, clear_request_ctx, set_research_run_id, get_logger
 from nsct.orchestration.budget import HardBudgetConfig
 from nsct.orchestration.state import ResearchRunState
+from nsct.providers.priority_queue import Priority
 from nsct.metrics import (
     metrics,
     C_SEARCH_QUERIES_TOTAL,
@@ -410,6 +411,7 @@ async def _run_pipeline(run: _ResearchRunState, request: ResearchRequest, budget
             query=request.query,
             budget_config=budget,
             depth=request.depth,
+            priority=Priority.NORMAL,
         )
 
         stage_ctx["stage"] = "planning"