System Information

OS: MacOS ARM, Android or Linux aarch64.
Appeared in #25915

Detailed description

1. vx_setall_f16 does not work with regular float constants. e.g. vx_setall_f16(-10.7421875f) leads to compiler error:

/Users/opencv-cn/GHA-OCV-3/_work/opencv/opencv/opencv/modules/core/include/opencv2/core/hal/intrin_math.hpp:45:40: error: no matching function for call to 'vx_setall_f16'
        const v_float16 _vexp_lo_f16 = vx_setall_f16(-10.7421875f);
                                       ^~~~~~~~~~~~~
/Users/opencv-cn/GHA-OCV-3/_work/opencv/opencv/opencv/modules/core/include/opencv2/core/hal/intrin.hpp:573:22: note: candidate function not viable: no known conversion from 'float' to 'cv::hfloat' for 1st argument

2. __test_exp((float16_t) 10, (float16_t) 1e-2, (float16_t) 1e2, flt16_min) cannot be build due to float16_t conflict with cv::hfloat.

Steps to reproduce

Remove disabler at test_intrin_utils.hpp:1954
Build for Mac M1

Issue submission checklist

• I report the issue, it's not a question
• I checked the problem with documentation, FAQ, open issues, forum.opencv.org, Stack Overflow, etc and have not found any solution
• I updated to the latest OpenCV version and the issue is still there
• There is reproducer code and related data files (videos, images, onnx, etc)