fix(nnUNet): Correct background mask in BraTS preprocessor

[Flink-ddd]

Hi, Maintainer.

This PR fixes the issue described in #1438.

Problem:
The preprocessing code for the BraTS nnU-Net example was using np.where(image[i] <= 0) to create a background mask. This incorrectly masked out foreground voxels that had negative values after z-score normalization.

Solution:
This fix changes the condition to np.where(image[i] == 0), which correctly identifies only the true background voxels. This ensures the 5th channel mask is generated as intended.

Fixes #1438

[Flink-ddd]

Hi @nv-kkudrynski,

I hope you are doing well.

I'm just gently checking in on this pull request, which aims to fix the issue #1438 regarding an incorrect background mask generation in the BraTS preprocessor.

Could you please take a look when you have a moment? Or perhaps suggest another reviewer who might be familiar with this part of the codebase?

Thank you for your time and for maintaining this great project!