fix: cannot be narrowed from 'double' to 'float'

src/include/cuda_math_helper.h

@@ -917,13 +917,13 @@ HD_FUNC double rnorm4d(double a, double b, double c, double d) {
     return {a.y * b.z - a.z * b.y, a.z * b.x - a.x * b.z, a.x * b.y - a.y * b.x};            \
   }                                                                                          \
   HD_FUNC T2 sqrt(T2 a) {                                                                    \
-    return {sqrt(a.x), sqrt(a.y)};                                                           \
+    return {sqrtf(a.x), sqrtf(a.y)};                                                           \
   }                                                                                          \
   HD_FUNC T3 sqrt(T3 a) {                                                                    \
-    return {sqrt(a.x), sqrt(a.y), sqrt(a.z)};                                                \
+    return {sqrtf(a.x), sqrtf(a.y), sqrtf(a.z)};                                                \
   }                                                                                          \
   HD_FUNC T4 sqrt(T4 a) {                                                                    \
-    return {sqrt(a.x), sqrt(a.y), sqrt(a.z), sqrt(a.w)};                                     \
+    return {sqrtf(a.x), sqrtf(a.y), sqrtf(a.z), sqrtf(a.w)};                                     \
   }                                                                                          \
   HD_FUNC T norm(T2 a) {                                                                     \
     return sqrt(dot(a, a));                                                                  \