GL: whoops, mapping of these to a TextureFormat was wrong.

Only noticed this now when adding inverse mapping. Sigh. OTOH, with the
inverse mapping in place this will no longer be possible to happen, as
it would cause a compile error due to a duplicate switch case.

doc/changelog.dox

@@ -993,6 +993,9 @@ See also:
     for most shader-related limits
 -   Fixed @ref GL::Shader reporting errors and warnings with line numbers
     off-by-one on desktop GLSL < 330.
+-   Fixed accidentally incorrect @ref GL::textureFormat() mapping for
+    @ref PixelFormat::RG8Snorm, @relativeref{PixelFormat,RGB8Snorm},
+    @relativeref{PixelFormat,RGBA8Snorm} and @relativeref{PixelFormat,RGB32UI}
 -   Fixed assertions related to OpenGL driver workarounds when the proprietary
     AMDGPU PRO drivers are used on Linux
 -   Fixed @relativeref{Corrade,Utility::Tweakable} parsers to compile with

src/Magnum/GL/Implementation/pixelFormatMapping.hpp

@@ -44,9 +44,9 @@ _n(RGBA8Unorm, RGBA, UnsignedByte)
 #endif
 #ifndef MAGNUM_TARGET_GLES2
 _c(R8Snorm, Red, Byte, R8Snorm)
-_c(RG8Snorm, RG, Byte, R8Snorm)
-_c(RGB8Snorm, RGB, Byte, R8Snorm)
-_c(RGBA8Snorm, RGBA, Byte, R8Snorm)
+_c(RG8Snorm, RG, Byte, RG8Snorm)
+_c(RGB8Snorm, RGB, Byte, RGB8Snorm)
+_c(RGBA8Snorm, RGBA, Byte, RGBA8Snorm)
 #else
 _s(R8Snorm)
 _s(RG8Snorm)
@@ -133,7 +133,7 @@ _c(RGB16I, RGBInteger, Short, RGB16I)
 _c(RGBA16I, RGBAInteger, Short, RGBA16I)
 _c(R32UI, RedInteger, UnsignedInt, R32UI)
 _c(RG32UI, RGInteger, UnsignedInt, RG32UI)
-_c(RGB32UI, RGBInteger, UnsignedInt, RGBA32UI)
+_c(RGB32UI, RGBInteger, UnsignedInt, RGB32UI)
 _c(RGBA32UI, RGBAInteger, UnsignedInt, RGBA32UI)
 _c(R32I, RedInteger, Int, R32I)
 _c(RG32I, RGInteger, Int, RG32I)