Refactoring ConvModule by removing norm_cfg

[sungchul2]

Summary

This PR includes:

• Update build_norm_layer

  • This function can handle various format of normalization:

    given format		returned format
    tuple[norm_name, norm_layer]	generated by build_norm_layer before	tuple[norm_name, nn.Module]	returned as is
    nn.Module	predefined module	("", nn.Module)	returned with empty name string
    nn.BatchNorm2d	type is only given	(norm_name, nn.Module)	built and returned with default settings and given num_features
    partial(nn.BatchNorm2d, eps=1)	partial for norm layer is given	(norm_name, nn.Module)	built and returned with custom settings and given num_features
    partial(build_norm_layer(nn.BatchNorm2d, requires_grad=True))	partial of build_norm_layer for arguments of build_norm_layer	(norm_name, nn.Module)	built and returned with custom settings and given num_features

• Replace norm_cfg with normalization

• Restore build_activation_layer

• Update ConvModule to use pre-instantiated normalization module and activation module

• Update activation docstring

How to test

Checklist

• I have added unit tests to cover my changes.​
• I have added integration tests to cover my changes.​
• I have ran e2e tests and there is no issues.
• 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) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

[sungchul2]

@sooahleex @harimkang I saw the performance drop for deit_tiny and efficientnet_v2 on hlabel classification as below.
I used only medium set (hlabel_CUB_medium), but classification has been using deterministic seed and its performance was maintained.

commit	deit_tiny	efficientnet_v2
1a8e10e	0.93	0.97
2ecaac1	0.87	0.39

If you already know this, feel free to ignore it :)

[eunwoosh]

Thanks for your lots of work :) LGTM. I left some minor comments. Please check it's worth to apply later.