drm/vmwgfx: Switch to mode_cmd2

Surprisingly few changes needed to make it happen. Compile-tested
only. The idea is that this replaces the 2 patches from Ville's big
fb->format patch series as a prep patch. Only impact to later patches
should be the one instace added in this patch where we look at
fb->pixel_format (instead of fb->bpp and fb->depth), so minor
adjustements in the cocci-generated patches needed.

v2: Restore pitch computation in vmw_fb_kms_framebuffer (Sinclair).

Cc: ville.syrjala@linux.intel.com
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: linux-graphics-maintainer@vmware.com
Cc: Sinclair Yeh <syeh@vmware.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161202070740.31689-1-daniel.vetter@ffwll.ch

drivers/gpu/drm/vmwgfx/vmwgfx_kms.h

@@ -248,7 +248,7 @@ vmw_kms_new_framebuffer(struct vmw_private *dev_priv,
 			struct vmw_dma_buffer *dmabuf,
 			struct vmw_surface *surface,
 			bool only_2d,
-			const struct drm_mode_fb_cmd *mode_cmd);
+			const struct drm_mode_fb_cmd2 *mode_cmd);
 int vmw_kms_fbdev_init_data(struct vmw_private *dev_priv,
 			    unsigned unit,
 			    u32 max_width,