Using std for argmin and argmax where possible

Author: tdegeus

include/xtensor/xsort.hpp

@@ -857,7 +857,10 @@ namespace xt
     {
         using value_type = typename E::value_type;
         auto&& ed = eval(e.derived_cast());
-        return detail::arg_func_impl<L>(ed, std::less<value_type>());
+        auto begin = ed.template begin<L>();
+        auto end = ed.template end<L>();
+        std::size_t i = std::distance(begin, std::min_element(begin, end));
+        return xtensor<size_t, 0>{i};
     }
 
     /**
@@ -884,7 +887,10 @@ namespace xt
     {
         using value_type = typename E::value_type;
         auto&& ed = eval(e.derived_cast());
-        return detail::arg_func_impl<L>(ed, std::greater<value_type>());
+        auto begin = ed.template begin<L>();
+        auto end = ed.template end<L>();
+        std::size_t i = std::distance(begin, std::max_element(begin, end));
+        return xtensor<size_t, 0>{i};
     }
 
     /**