This commit fixes the issue caused by #357 that nanovg failed to compile on iOS due to the wrong way of including glext header.

- added NVG_DEBUG which can be used to omit calls to glGetError()
- removed return value for glnvg__checkError
- create flag defines to enum

- all colors in the shader is handled as premultiplied
- removed alphaBlend param from nvgBeginFrame(), it became obsolete
- added NVGL_TEXTURE_PREMULTIPLIED flag
- separated FBO example
- FBOs are treated as premultipied for correct alpha
- changed FBO storage from depth/stencil 24/8 to stencil 8
- moved discard to end of shader which restores the render speed again

- added (optional) 3-pass stencil trick to stroke rendering, reduces
overdraw artifacts, a little slower
- fixed some text bounds issues

- fixed flags setup in nvglCreateImageFromHandle
- changed frame buffer to only compile on GL3
- changed nvgluCreateFramebuffer() to return FBO struct instead of
filling existing one (for consistency)

- added documentation about changed GL state
- made old GL2 and GL3 backends obsolete

- added support for GL(ES)2 for nanovg_gl3buf (something is still not
quite there)
- nanovg now sets all required GL state
- removed compiler warnings about unused vars

- added support to take screen shots
- added support to render to frame buffer using premultiplied alpha
- fixed spelling NVGaling -> NVGalign

- Renamed FPScounter to PerfGraph
- dump avg times at exit

- moved performance counter and drawing to separate file
- tweaks to performance drawing

- added NVG_NOTUSED to suppress unused param warnings
- fixed scissor anti-aliased border width when zoomed

- fixed off-by-one pixels from demo
- added device-pixel-ratio support (retina support)
- added nvgEndFrame()

- use nvgCreateGLES2() instead of nvgCreateGL2() when NANOVG_GLES2 is
specified (ditto for gl3)
- use GL_RED instead of GL_LUMINANCE in ES3

- added option to turn off AA fringe generation
- added examples which use no AA fringes and  MSAA

- merged GL ES2 and GL2 backends
- GL2 uses now same structure as GL3

- added milliseconds to fps counter
- smooth fps reading a bit more
- changed GL3 to uber shader

- set swap interval to 0 to better see performance
- draw convex shapes without stencil to speed things up
- improve vertex buffer upload for gl3

- moved demo stuff to separate file
- created two separate examples for gl2 and gl3
- initial stab at OGL 3.2 core profile back-end (currently twice as
slows as 2.0)
- API change: glBeginFrame() takes view witdth and height as input
- added FPS counter to demo