mp7 seg faults if first stress period, first time step arent in the headsave file

[jtwhite79]

It seems that for structured grids with the mf2k5 family of flow models, if you dont write the first stress period and first time step to the headsave file, mp7 seg faults somewhere in the ibound processing bc the particle tracking engine instance didnt get its grid attribute initialized, which I think is happening because the head reader instance doesnt have its layer and cellcount set properly (I think this is what is happening - took me a heap of fumbling with gdb dig this one out!). See the attached example...

mp7_ex1_mfnwt_dis.zip

[aprovost-usgs]

@jtwhite79 Thanks for reporting this issue, and especially for diving in with the debugger. We'll look at your example and make mp7 check for this condition and die more gracefully