Cleanup loggers to prevent double-logging

[mgoin]

This has bugged me for so long :)

I tested with deepsparse.benchmark using a SparseZoo stub

Before:

deepsparse.benchmark zoo:cv/detection/yolov5-n/pytorch/ultralytics/coco/base_quant-none
2023-06-06 14:46:45 deepsparse   INFO     Thread pinning to cores enabled
DeepSparse, Copyright 2021-present / Neuralmagic, Inc. version: 1.6.0 COMMUNITY | (3cd0b1a4) (optimized) (system=avx512, binary=avx512)
2023-06-06 14:46:47 deepsparse   INFO     deepsparse.engine.Engine:
        onnx_file_path: /home/mgoin/.cache/sparsezoo/42de2f1d-3a2c-4445-aa94-63aa12e1e2a5/model.onnx
        batch_size: 1
        num_cores: 18
        num_streams: 1
        scheduler: Scheduler.default
        fraction_of_supported_ops: 1.0
        cpu_avx_type: avx512
        cpu_vnni: False
INFO:deepsparse:deepsparse.engine.Engine:
        onnx_file_path: /home/mgoin/.cache/sparsezoo/42de2f1d-3a2c-4445-aa94-63aa12e1e2a5/model.onnx
        batch_size: 1
        num_cores: 18
        num_streams: 1
        scheduler: Scheduler.default
        fraction_of_supported_ops: 1.0
        cpu_avx_type: avx512
        cpu_vnni: False
2023-06-06 14:46:47 deepsparse.utils.onnx INFO     Generating input 'images', type = float32, shape = [1, 3, 640, 640]
INFO:deepsparse.utils.onnx:Generating input 'images', type = float32, shape = [1, 3, 640, 640]
2023-06-06 14:46:47 deepsparse   INFO     Starting 'singlestream' performance measurements for 10 seconds

After

deepsparse.benchmark zoo:cv/detection/yolov5-n/pytorch/ultralytics/coco/base_quant-none
2023-06-06 14:56:47 deepsparse.benchmark.benchmark_model INFO     Thread pinning to cores enabled
DeepSparse, Copyright 2021-present / Neuralmagic, Inc. version: 1.6.0 COMMUNITY | (3cd0b1a4) (optimized) (system=avx512, binary=avx512)
2023-06-06 14:56:49 deepsparse.benchmark.benchmark_model INFO     deepsparse.engine.Engine:
        onnx_file_path: /home/mgoin/.cache/sparsezoo/neuralmagic/yolov5-n-coco-base_quantized/model.onnx
        batch_size: 1
        num_cores: 18
        num_streams: 1
        scheduler: Scheduler.default
        fraction_of_supported_ops: 1.0
        cpu_avx_type: avx512
        cpu_vnni: False
2023-06-06 14:56:49 deepsparse.utils.onnx INFO     Generating input 'images', type = uint8, shape = [1, 3, 640, 640]
2023-06-06 14:56:49 deepsparse.benchmark.benchmark_model INFO     Starting 'singlestream' performance measurements for 10 seconds

[rahul-tuli]

This was such a good find