radeonsi: always prefer SWITCH_ON_EOP(0) on CIK

radeonsi: always prefer SWITCH_ON_EOP(0) on CIK

The code is rewritten to take known constraints into account, while always
using 0 by default.

This should improve performance for multi-SE parts in theory.

A debug option is also added for easier debugging. (If there are hangs,
use the option. If the hangs go away, you have found the problem.)

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
v2: fix a typo, set max_se for evergreen GPUs according to the kernel driver