change ReatT to Double in generated api

[piiertho]

This remove use of RealT in favor of Double in generated API code, according to #185
I am not able to test it right now as I encounter linking error on OSX, when compiling samples:

  "_godot_color_set_r", referenced from:
      _godot_gdnative_godot_color_set_r_wrapper3 in libgodot-library-cinterop-gdnative-cache.a(result.o)
     (maybe you meant: knifunptr_godot_gdnative3_godot_color_set_r, _godot_gdnative_godot_color_set_r_wrapper3 )
  "_godot_string_to_int64", referenced from:
      _godot_gdnative_godot_string_to_int64_wrapper691 in libgodot-library-cinterop-gdnative-cache.a(result.o)
     (maybe you meant: knifunptr_godot_gdnative691_godot_string_to_int64, _godot_gdnative_godot_string_to_int64_wrapper691 )
  "_godot_string_char_uppercase", referenced from:
      _godot_gdnative_godot_string_char_uppercase_wrapper708 in libgodot-library-cinterop-gdnative-cache.a(result.o)
     (maybe you meant: knifunptr_godot_gdnative708_godot_string_char_uppercase, _godot_gdnative_godot_string_char_uppercase_wrapper708 )
  "_godot_pool_string_array_invert", referenced from:
      _godot_gdnative_godot_pool_string_array_invert_wrapper232 in libgodot-library-cinterop-gdnative-cache.a(result.o)
     (maybe you meant: knifunptr_godot_gdnative232_godot_pool_string_array_invert, _godot_gdnative_godot_pool_string_array_invert_wrapper232 )
  "_godot_quat_new_with_euler", referenced from:
      _godot_gdnative_godot_quat_new_with_euler_wrapper73 in libgodot-library-cinterop-gdnative-cache.a(result.o)
     (maybe you meant: knifunptr_godot_gdnative73_godot_quat_new_with_euler, _godot_gdnative_godot_quat_new_with_euler_wrapper73 )
ld: symbol(s) not found for architecture x86_64

I have the same behaviour with master-merge branch.
Can you test it on Linux and Windows ? As I am away from home I only have a macbook with me.

[chippmann]

I tested on Linux and there i have no problem.
But if I can test on MacOs this evening.
But i guess it does not affect this PR. The fix would have to be together with the issue you created for it.

[piiertho]

I tested on Linux and there i have no problem.
But if I can test on MacOs this evening.
But i guess it does not affect this PR. The fix would have to be together with the issue you created for it.

Yep, this the osx link issue is not linked to this PR. I just mentionned it as I can't test samples from osx.