Use GL_QUADS for flushing a quad batch

Instead of using GL_TRIANGLES and uploading the indices every time, it
now uses GL_QUADS instead on OpenGL. Under GLES it still uses indices
but it uses the new cogl_vertex_buffer_indices_get_for_quads function
to avoid uploading the vertices every time.

This requires the _cogl_vertex_buffer_indices_pointer_from_handle
function to be exposed privately to the rest of Cogl.

The static_indices array has been removed from the Cogl context.

gles/cogl-context.c

@@ -72,7 +72,6 @@ cogl_create_context ()
 
   _context->journal = g_array_new (FALSE, FALSE, sizeof (CoglJournalEntry));
   _context->logged_vertices = g_array_new (FALSE, FALSE, sizeof (GLfloat));
-  _context->static_indices = g_array_new (FALSE, FALSE, sizeof (GLushort));
   _context->polygon_vertices = g_array_new (FALSE, FALSE,
                                             sizeof (CoglTextureGLVertex));
 
@@ -162,8 +161,6 @@ _cogl_destroy_context ()
   if (_context->logged_vertices)
     g_array_free (_context->logged_vertices, TRUE);
 
-  if (_context->static_indices)
-    g_array_free (_context->static_indices, TRUE);
   if (_context->polygon_vertices)
     g_array_free (_context->polygon_vertices, TRUE);
   if (_context->current_layers)

gles/cogl-context.h

@@ -80,7 +80,6 @@ typedef struct
    * can batch things together. */
   GArray           *journal;
   GArray           *logged_vertices;
-  GArray           *static_indices;
   GArray           *polygon_vertices;
 
   /* Some simple caching, to minimize state changes... */