MSVC compile compatible (#13)

* to_string() and from_string()

Added to_string and from_string methods that are not related to floating point numbers

* fixed_string unit_test updated.

Generated by the Cursor, checked by me. I simply run it in MSVC and no error is found.
I am very glad that you replied so fast. Under my test, your lib is fast enough for a gaming environment, except a little bit slow on division. But there is some critical functions are missing:
mod
exp (power)
log
printf scanf (solved)
hyperbolic sine

* bug found and correct

* that is right.

* upgrade tests, fix constexpr in code

* MSVC compile compatible

When compiled in MSVC (VS 2022), those places above will lead to a compile failure.

---------

Co-authored-by: Artur Bać <artur@ebasoft.com.pl>

Author: UF4007

fixed_lib/include/fixedmath/math.h

@@ -819,12 +819,12 @@ namespace detail
     {
     fixed_internal x2{mul_<prec_>(x, x)};
 
-    constexpr fixed_internal _35o9{div_<prec_>(35, 9) + 1};
-    constexpr fixed_internal _5o7{div_<prec_>(5, 7) + 1};
-    constexpr fixed_internal _3o5{div_<prec_>(3, 5) + 1};
-    constexpr fixed_internal _1o3{div_<prec_>(1, 3)};
-    constexpr fixed_internal _1{fix_<prec_>(1)};
-    constexpr fixed_internal _63o11{div_<prec_>(63, 11) + 1};
+    constexpr fixed_internal _35o9{div_<prec_>(fixed_internal{35}, fixed_internal{9}) + 1};
+    constexpr fixed_internal _5o7{div_<prec_>(fixed_internal{5}, fixed_internal{7}) + 1};
+    constexpr fixed_internal _3o5{div_<prec_>(fixed_internal{3}, fixed_internal{5}) + 1};
+    constexpr fixed_internal _1o3{div_<prec_>(fixed_internal{1}, fixed_internal{3})};
+    constexpr fixed_internal _1{fix_<prec_>(fixed_internal{1})};
+    constexpr fixed_internal _63o11{div_<prec_>(fixed_internal{63}, fixed_internal{11}) + 1};
 
     fixed_internal y6{_35o9 + mul_<prec_ + 1>(x2, _63o11)};
     fixed_internal y7{_5o7 + mul_<prec_ + 3>(x2, y6)};

perf_test_suite/perf.cc

@@ -7,6 +7,7 @@
 #include <chrono>
 #include <cmath>
 #include <string_view>
+#include <numeric>
 using std::chrono::duration_cast;
 using std::chrono::microseconds;
 using std::chrono::milliseconds;
@@ -22,6 +23,8 @@ using std::tan;
 using std::sin;
 
 using fixedmath::fixed_t;
+using fixedmath::phi;
+using namespace fixedmath::func;
 
 using limits_ = std::numeric_limits<fixed_t>;