Refactor inference engine and task management: Remove obsolete inference engine and ONNX engine files, update inference processing to utilize batch handling, and enhance task management structure in documentation. Adjust paths for task specifications to align with new directory organization.

e2e/tests/test_resource_limits.py

@@ -10,16 +10,14 @@ from pathlib import Path
 import pytest
 
 
-def _video_ai_body(mock_loader_url: str, video_rel: str) -> dict:
-    base = mock_loader_url.rstrip("/")
-    name = video_rel.rstrip("/").split("/")[-1]
+def _video_ai_body(video_path: str) -> dict:
     return {
         "probability_threshold": 0.25,
         "tracking_intersection_threshold": 0.6,
         "altitude": 400,
         "focal_length": 24,
         "sensor_width": 23.5,
-        "paths": [f"{base}/load/{name}"],
+        "paths": [video_path],
         "frame_period_recognition": 4,
         "frame_recognition_seconds": 2,
     }
@@ -31,12 +29,11 @@ def test_ft_n_08_nft_res_lim_02_sse_queue_bounded_best_effort(
     warm_engine,
     http_client,
     jwt_token,
-    mock_loader_url,
     video_short_path,
     sse_client_factory,
 ):
     media_id = f"rlim-sse-{uuid.uuid4().hex}"
-    body = _video_ai_body(mock_loader_url, video_short_path)
+    body = _video_ai_body(video_short_path)
     headers = {"Authorization": f"Bearer {jwt_token}"}
     collected: list[dict] = []
     thread_exc: list[BaseException] = []