Fix a performance drop while training EfficientNetV2 with multi-GPU

[eunwoosh]

Summary

This PR fixes #2253
The reason of performance drop is not to use a function converting BatchNorm to SyncBatchNorm from timm.
There is additional code in timm's function to deal with some cases in timm's model, so I should have used that function for models from timm like EfficientNetV2.
So, I changed code to use timm's function for timm's model.

How to test

Checklist

• I have added unit tests to cover my changes.​
• I have added integration tests to cover my changes.​
• I have added e2e tests for validation.
• I have added the description of my changes into CHANGELOG in my target branch (e.g., CHANGELOG in develop).​
• I have updated the documentation in my target branch accordingly (e.g., documentation in develop).
• I have linked related issues.

License

• I submit my code changes under the same Apache License that covers the project.
  Feel free to contact the maintainers if that's a concern.
• I have updated the license header for each file (see an example below).

# Copyright (C) 2023 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

[goodsong81]

LGTM!