[Testing] Rewrite python tests for Normalize, CvtColor, BilateralFilter, BndBox, Histogram, GammaContrast, Composite

[zacharyvincze]

Motivation

Aims to rewrite python tests so that they are no longer reliant on expected test images.

Technical Details

• Rewrites python tests for Normalize, CvtColor, BilateralFilter, BndBox, Histogram, GammaContrast and Composite.
• Fixes a bug in the Histogram python operator definition for which the batch size was hard-coded to 1 for the generated output tensor size.
• Removes the now unused expected test images.

Test Plan

• Run python/C++ tests to ensure new tests pass and that there is no regression (with the removal of test images).

Test Result

• All C++/Python tests are passing as expected.

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[Copilot]

Pull Request Overview

This PR rewrites Python tests for multiple computer vision operators to eliminate dependency on expected test images and makes tests more maintainable by using generated random tensors instead. It also fixes a batch size bug in the Histogram operator's Python bindings.

• Rewrites tests for Normalize, CvtColor, BilateralFilter, BndBox, Histogram, GammaContrast, and Composite operators
• Fixes hardcoded batch size bug in Histogram operator Python definition
• Removes dependency on expected test images by using tensor comparison approach

Reviewed Changes

Copilot reviewed 10 out of 28 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
test_op_normalize.py	Replaces image-based tests with parametrized tests using generated tensors
test_op_non_max_suppression.py	Updates to use tensor generation and comparison instead of binary file comparison
test_op_histogram.py	Converts from grayscale image input to generated tensor input with tensor comparison
test_op_gamma_contrast.py	Replaces image loading with generated tensor approach and tensor comparison
test_op_cvt_color.py	Updates color conversion tests to use generated tensors and parametrized color codes
test_op_composite.py	Converts composite operation tests to use generated tensors for all inputs
test_op_bnd_box.py	Replaces hardcoded bounding box setup with random box generation
test_op_bilateral_filter.py	Updates bilateral filter tests with generated tensors and expanded parameter testing
test_helpers.py	Adds generic tensor generation function and updates parameter naming
py_op_histogram.cpp	Fixes batch size calculation bug that was hardcoded to 1

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}