Remove [Engine] and [EnginePoolManager] debug log messages

Cleaned up verbose engine telemetry emitted to stdout/stderr and the
Windows Event Viewer. Removes logEngineEvent/logEvent calls (and their
diagnostic-only locals) across the TensorRT engine load, build, run,
multi-GPU, and pool-manager paths, plus the now-unused logEvent helper
in EnginePoolManager.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

engines/TensorRTAPI/include/engine/EngineMultiGpu.inl

@@ -223,9 +223,6 @@ bool Engine<T>::loadSlots(
         : probeEngine->loadNetwork     (modelPath, subVals, divVals, normalize);
 
     if (!probeOk) {
-        logEngineEvent("[Engine] loadSlots FAIL: Probe engine failed on GPU["
-            + std::to_string(probeGpuIdx) + "] for " + modelPath
-            + " (freeVRAM before=" + std::to_string(freeBefore / 1048576) + " MiB)", true);
         return false;
     }
 
@@ -686,13 +683,6 @@ bool Engine<T>::runInferenceFromPool(
     }
     if (!slot) {
         ANS_DBG("TRT_Pool", "ERROR: No slot available — all %zu slots busy, timeout", m_slots.size());
-        std::string errMsg = "[Engine] runInferenceFromPool FAIL: Capacity reached -- all "
-            + std::to_string(m_activeCount.load()) + "/" + std::to_string(m_totalCapacity)
-            + " slot(s) busy"
-            + (m_elasticMode ? " and all GPUs full" : "")
-            + ". Request rejected (2s timeout).";
-        std::cout << errMsg << std::endl;
-        logEngineEvent(errMsg, true);
         return false;
     }