This function takes `struct mrb_rational*` and returns converted
`mrb_float` value.

This change relies that `mrb_num_div_flo` does not use `mrb` inside.

This function handles zero division properly. Also fixed bugs that multiply
numbers instead of division.

Use `NaN*i` as CRuby does.

Use global defines for `mruby-bin-debugger`

Reorganize types for `ObjectSpace.count_objects`

#### Addition

- T_COMPLEX
- T_RATIONAL

#### Deletion

- T_FALSE
- T_FREE
- T_TRUE
- T_SYMBOL
- T_UNDEF

Add example for cross-compiling to Nintendo Switch

Add configuration for MinGW cross compilation [ci skip]

along with repeated calls of `strlen()`.

Cross Compiling configuration for the Nintendo Switch, it requires Nintendo SDK

With this cross compiler configuration, all tests for full-core gembox on
Windows on GitHub Actions were successful.

This change also fixes the error caused by `rational.c` that calls
`mrb_complex_eq()`, which had been undefined.

`MRB_INT_MIN` does not have a corresponding positive value.

- define `MRB_TT_COMPLEX`
- change object structure (`struct RComplex`)
- add memory management for `MRB_TT_COMPLEX`
- avoid operator overloading as much as possible
- as a result, performance improved a log
- should work with and without `Rational` defined

- define `MRB_TT_RATIONAL`
- change object structure (`struct RRational`)
- add memory management for `MRB_TT_RATIONAL`
- avoid operator overloading as much as possible
- implement division overloading in C
- as a result, performance improved a lot

Since `mruby` does not have `Bignum`, `Float#divmod` could overflow, so
it will return `Float` values when the divided value does not fit in
`mrb_int`. This behavior will be changed when `Bignum` is introduced to
`mruby` in the future.

`Float::NAN/0` should be `Float::NAN`; ref a0b3378b

#### Before this patch:

```console
$ bin/mruby -e 'p(Float::NAN/0)'
Infinity
```

#### After this patch (same as Ruby):

```console
$ bin/mruby -e 'p(Float::NAN/0)'
NaN
```