glgsg: use T_float when downloading 24-bit depth textures

This is to match behavior with previous Panda3D versions.  See #212

panda/src/glstuff/glGraphicsStateGuardian_src.cxx

@@ -6460,9 +6460,6 @@ framebuffer_copy_to_ram(Texture *tex, int view, int z,
     break;
 
   case Texture::F_depth_component24:
-    component_type = Texture::T_unsigned_int;
-    break;
-
   case Texture::F_depth_component32:
     component_type = Texture::T_float;
     break;