Fix test_correctness_abs_float64

[zasdfgbnm]

Fixes #4852

[zasdfgbnm]

!test

[github-actions]

Description

• Added abs function for double precision

• Ensured consistent absolute value computation

---

Changes walkthrough 📝

	Relevant files
Enhancement	helpers.cu Added double precision `abs` function                                        runtime/helpers.cu Added abs function for double precision Ensured consistent absolute value computation +4/-0

---

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

🧪 No relevant tests

⚡ Recommended focus areas for review Possible Issue The abs function for double type is implemented manually instead of using the standard library function fabs. This could lead to inconsistencies or potential issues with precision and performance. __device__ double abs(double a) { return a < 0 ? -a : a; }

[naoyam]

Not sure why this was closed (just accidentally?). @zasdfgbnm, could you please finish fixing the issue?