Blender Git Loki

Shifted the dispatch from the algorithm side to the storage side of the
sampling system to keep the code simple.

Now there is a MSurfaceSampleStorage struct that encodes the storage
details, which the algorithms don't have to care about.

Added generalized method for creating an array of mesh samples with
arbitrary stride.

Changed mesh sample definition to use 3 vertex weights instead of a
face index. This is easier to sample uniformly and avoids the need for
tesselation for evaluating.

Normalize the normal vector after sample eval to account for
interpolation.

Added a bool return to the eval function to give feedback on invalid
samples.

Added a basic RNA definition for the mesh sampling system.

Eval function to get a location and normal vector from mesh samples.

Usable random distribution algorithm.

Does not include area weighting yet.

Skeleton code for sampling meshes.

Fix one more issue - we need a string length callback too, for pure 'virtual' RNA strings...

Fix a mem leak in file listing threaded code, and another bug in fsmenu.c (new named bookmarks).

Merge branch 'master' into asset-experiments

Conflicts:
release/scripts/startup/bl_ui/__init__.py
source/blender/editors/interface/interface_utils.c

Merge branch 'master' into mesh-transfer-data

Conflicts:
source/blender/makesrna/intern/rna_modifier.c

Revert "Sticky keys branch squashed commit"

This reverts commit d3a3e4a52cab4a9ff4d444f0788dcd6e4e62aecd.

Sticky keys branch squashed commit

Merge remote-tracking branch 'origin/master' into experimental-build

Vertex painting PBVH refactor:

Refactor PBVH off SculptSession and in a separate PaintSession struct
Entering Vertex painting will generate a PBVH and vertex painting now
draws using PBVH.
Add stub vertex painting operator that will use PBVH.

Depsgraph: Implement real node layers flush

It's not optimal at this moment, to optimize such a traversal we'd need to have
some ID level dependencies, so we can query much faster which IDs depends on
which else IDs.

Merge branch 'master' into depsgraph_refactor

cleanup diffusion dof names and add blade count (will possibly be used in high quality)