doc: various fixes.

Thx @biosek.

doc/best-practices.dox

@@ -77,11 +77,11 @@ vertex and index buffers in both Native Client and Emscripten.
 
 @subsection best-practices-powervr PowerVR hardware
 
-- [PowerVR Performance Recommendations](http://www.imgtec.com/powervr/insider/docs/PowerVR.Performance%20Recommendations.1.0.28.External.pdf) [PDF]
+- [PowerVR Performance Recommendations](https://github.com/burningsun/pecker_framework/blob/master/参考资料/PowerVR.Performance%20Recommendations.pdf?raw=true) [PDF]
 
 @subsection best-practices-tegra NVidia Tegra hardware
 
-- [Optimize OpenGL ES 2.0 Performance for Tegra](http://docs.nvidia.com/tegra/data/Optimize_OpenGL_ES_2_0_Performance_for_Tegra.html)
+- [Optimize OpenGL ES 2.0 Performance for Tegra](http://docs.nvidia.com/tegra/index.html#GLES2_Perf_Main.html)
 
 */
 }

doc/platform.dox

@@ -190,7 +190,7 @@ constructor. Using method chaining it can be done conveniently like this:
 MyApplication::MyApplication(int& argc, char** argv):
     Platform::Application(argc, argv, Configuration()
         .setTitle("My Application")
-        .setSize({800, 600})
+        .setSize({800, 600}))
 {
     // ...
 }

doc/transformations.dox

@@ -154,7 +154,7 @@ auto b = Matrix4::reflection(axis.normalized());
 Scaling and reflection is also done relative to origin, you can use method
 mentioned above to scale or reflect around arbitrary point.
 
-Sscaling and reflection can be (to some extent) also represented by complex
+Scaling and reflection can be (to some extent) also represented by complex
 numbers and quaternions, but it has some bad properties and would make some
 operations more expensive, so it's not implemented.