this sounds right, the coordinate system printed should match the mpu coordinate system. Not sure about the mag though...
Did you already have a chance to test the imu_drotek branch?
You should be able to have a rather empty imu section (except mag calib, body_to_imu and accel neutrals if you want).
I got rid of the unused parameter warnings, not sure what to do about the cast warning yet...
On a side note: if you don't want to use GYRO_P_Q, etc. you can also just leave them out (so it won't even be calculated) instead of setting them to zero.