Add support for high priority scheduling in amdgpu v6

[Andres Rodriguez]

Updated with Christian and Alex's feedback.

Also added missing initialization of ring->num_jobs[] to PATCH 18

Added a new patch:
[PATCH 19/22] drm/amdgpu: convert srbm lock to a spinlock

Also some new benchmarks for the high priority queue using a modified
computeparticles that only takes 0.2ms to execute:
https://plot.ly/~lostgoat/15/

The 0.2ms time was picked as an estimate of how long the reprojection
task would take to execute.