Occlusion Culling

**Is your feature request related to a problem? Please describe.**
Objects that can't been seen are still sent to the GPU for rendering but to no visual effect. Lowering the computation costs of occluded objects improves application's performance, but with no impact to the scene quality.

**Describe the solution you'd like**
Software Occlusion Culling based on CPU rasterization of the depth buffer with axis-aligned bounding box tests to determine if an object is occluded.

Hardware Occlusion Culling to cull dynamic geometry in real-time by using a hierarchical depth buffer (generated by the GPU during a depth pre-pass) combined with GPU occlusion queries.

Precomputed Occlusion Culling based on visibility volumes where culling is performed at runtime by checking the cell the camera is currently inside, and retrieving a list of non-occluded meshes.

**OS and backend**
Software and Precomputed Occlusion Culling should be available on all platforms (except SIMD-optimized parts).

Hardware Occlusion Culling is limited to GPU models that supports occlusion queries.