render one frame per thread for increased parallelism

If the nodes are now stateless, there's nothing stopping the renderer from
rendering multiple frames at once. Earlier since the nodes held some of their
input/output data (and that data could change per frame), it was not possible
to render multiple frames at once without conflicts. Now that the node state is
held in render threads, they can do whatever they want at any time.

app/render/backend/videorenderbackend.cpp

@@ -43,6 +43,8 @@ void VideoRenderBackend::InvalidateCache(const rational &start_range, const rati
     return;
   }
 
+  RenderBackend::InvalidateCache(start_range, end_range);
+
   // Adjust range to min/max values
   rational start_range_adj = qMax(rational(0), start_range);
   rational end_range_adj = qMin(SequenceLength(), end_range);
@@ -185,11 +187,6 @@ VideoRenderFrameCache *VideoRenderBackend::frame_cache()
   return &frame_cache_;
 }
 
-ColorProcessorCache *VideoRenderBackend::color_cache()
-{
-  return &color_cache_;
-}
-
 const char *VideoRenderBackend::GetCachedFrame(const rational &time)
 {
   last_time_requested_ = time;
@@ -237,9 +234,3 @@ NodeInput *VideoRenderBackend::GetDependentInput()
 {
   return viewer_node()->texture_input();
 }
-
-void VideoRenderBackend::CompletedFrame()
-{
-  caching_ = false;
-  CacheNext();
-}