ppStack segv when using recipe STACK_1DG

[bills]

stack 2641329 and others die with fault 4
The error is in the function psMMGMM

i - offsets->data.S32[i] < 0 which leads to access outside the bounds of an array

for (int i = 0; i < N - Ncensored; i++) {

modes->data.F32[i - offsets->data.S32[i]] = tempModes->data.F32[i];
for (int j = 0; j < dim; j++) {

P->data.F32[i - offsets->data.S32[i]][j] = tempP->data.F32[i][j];

}

}

If the block of code is skipped when the invalid offset is detected
a memory corruption is detected later indicating that other
bad things may be going on.

[watersc1]

v36180 of psMixtureModels.c contains the fix for this.

I had the direction of the offsets reversed: "i - offsets[i]" will go negative, whereas "i + offsets[i]" properly places values where they belong.

This revision also contains a calculation fix for the initial population fractions. It was coming up with [1.0 0.0] for the initial fractions, and then failing to fit the second mode. I've reset this to be [0.5 0.5], which may be slightly slower (the first EM iteration will generate a better fraction, which the previous calculation was supposed to do automatically), but converges to the correct values.