Optimized Anim Graph by enabling Fast Path

[AHGS-Jens]

Anim Graph uses fast path instead of running blueprit logic if it is able to access variables directly. So in order to enable this I made wrappers for the enums used by the Anim Graph that updates bool values for each state of the enum that can be used to check if it is in a specific state.

In addition some structs was brought out of the Grounded and AimingValues structs. as Fast Track does not work when accessing structs within structs.

The net effect of this should be improved performance which I tried to measure using stat anim with 25 character constantly running around.

Getting accurate numbers is a bit difficult as performance seem to fluctuate quite a bit, but some of the gains are considerable enough that it certainly isn't down to margin of error.

	Before ms	After ms	Difference ms	Percent change
PerformAnimEval	11,10	10,39	-0,71	-6,40%
ComponentAnimEval	7,09	6,91	-0,18	-2,54%
EvaluateAnimInst	7,00	6,81	-0,19	-2,71%
EvaluateAnimGraph	6,99	6,81	-0,18	-2,58%
BlendPosesInGraph	6,97	6,78	-0,19	-2,73%
ProxyUpdate	3,51	2,97	-0,54	-15,38%
EvaluateAnimState	3,34	3,21	-0,13	-3,89%
StateMachineUpdate	2,11	1,64	-0,47	-22,27%
AnimGameThreadTime	2,11	2,1	-0,01	-0,47%
AnimDecompression	1,85	1,8	-0,05	-2,70%
AnimSeqGetBonePose	1,67	1,71	0,04	2,40%
BlendSpaceGetAnimPose	1,70	1,61	-0,09	-5,29%
PostAnimEval	1,17	1,14	-0,03	-2,56%
ExtractPoseFromAnim	1,17	1,13	-0,04	-3,42%
StateMachineFindTrans	1,12	0,95	-0,17	-15,18%
SkinnedMeshCompTick	0,94	0,97	0,03	3,19%
(N) LocomotionCycles	0,79	0,48	-0,31	-39,24%
UpdateKinematicBones	0,92	0,71	-0,21	-22,83%
AimOffset	0,63	0,64	0,01	1,59%
AnimTickTime	0,67	0,65	-0,02	-2,99%
BlendPosesPerBone	0,62	0,63	0,01	1,61%
UpdateAnimation	0,66	0,64	-0,02	-3,03%
NativeUpdateAnimation	0,56	0,52	-0,04	-7,14%
UpdateRBJointsMotors	0,47	0,51	0,04	8,51%

[dyanikoglu]

That was something I did in my private repo, but didn't have enough time to migrate it here. I will review your code and probably merge it this weekend. Sorry for the delay, and thanks for the contribution!

[dyanikoglu]

@AHGS-Jens Everything looks fine to me, can you solve the merge-conflict before I can merge this?

[AHGS-Jens]

Resolved the conflict and gave it a spin in the demo level, seems to work fine, but haven't done any further testing than that.