Update hcp_fix to support the MotionRegression=FALSE condition

[mharms]

With the forthcoming FIX version "1.06.12", we can now properly implement HP filtering in the MotionRegression=FALSE condition within single-run FIX. This PR checks for the FIX version, and calls fix appropriately based on the FIX version.

This should resolve #108.

[glasserm]

Has this been tested using the new options?

[mharms]

Yes, both hp=-1 and hp=2000 run to completion with doMotionRegression=FALSE.

[glasserm]

Does hp=0 work and do a detrend?

[mharms]

No, that's a completely separate issue. Support for hp=0 was not added, so the script still aborts if hp=0 is requested.

[glasserm]

Is that worth doing too, or a bigger job?

[mharms]

It's worth doing, but it is also a bigger job that I'm not going to tackle right now. We would either need to: (1) Implement an fsl_glm detrending of the volume data in the hp=0 condition, or (2) accomplish the filtering using functionhighpassandvariancenormalize (i.e., as we do in hcp_fix_multi_run).

[glasserm]

Thanks Mike