Deploy latest changes from main branch

src/core/services/ml_inference.py

@@ -207,7 +207,7 @@ async def run_inference(image_bytes: bytes) -> Dict[str, Any]:
             image_size_bytes=len(image_bytes),
             latency_ms=cached["latency_ms"],
         )
-        # await asyncio.sleep(45)  # TEMP: remove before production
+        await asyncio.sleep(20)  # TEMP: remove before production
         return cached
 
     logger.info(
@@ -233,7 +233,7 @@ async def run_inference(image_bytes: bytes) -> Dict[str, Any]:
             confidence=result["confidence"],
         )
         await redis_manager.set_cache(image_hash, result, ttl=settings.prediction_cache_ttl_s)
-        # await asyncio.sleep(45)  # TEMP: remove before production
+        await asyncio.sleep(20)  # TEMP: remove before production
         return result
     except (InvalidImageError, UnsupportedImageFormatError, InferenceServiceError):
         raise