* Instead of only doing a very lightweight check to see if the API is available
up-front, we now share the API initialisation among all API tests far enough
to determine availability of extensions, features, etc. Then we can precisely
determine which tests are available and which aren't before running.
* This is primarily around initial states - either a resource which is from a
previous frame and maybe wasn't dirtied and needs initial states created for
it when it's modified mid-frame, or a resource that's created and destroyed
all within one frame.
* We also pick the output pixel in the CBuffer_Zoo tests to ensure the API
agrees with our interpretation of the data.
* Follow-up commit will tidy D3D cbuffer code that needs it.
* If we create in a bare pointer then just cast to a smart pointer, it adds a
ref instead of stealing it. Just only allow creating smart pointers in the
first place.
* This is kind of a "damned if you do, damned if you don't" situation. If the
cullmode is wrong, having it show up as failures in depth/stencil overlays is
misleading. However if the cullmode is deliberate setting it to no-cull may
ruin the results of those overlays.
* The original behaviour of leaving cullmode as is ends up being a better
tradeoff, since at least when it is misleading you can use the culling overlay
to show the failure there. If culling is disabled it makes depth/stencil
overlays useless in situations with no recourse.
* Also add a test of a depth-clipped triangle to overlay tests
* Built entirely independently of the main renderdoc build (for now).
* Contains python scripts & framework for running tests on a renderdoc build,
which will be run nightly.
baldurk