It is possible that doping is not enought to change the mean characteristic of the matrix compound. Moreover, because Ni is not strong ferromagnet. Try to amplify the region near H=0 and seeing if there is a little (may be very little) magnetic component (deviation of the linear behavior).