stack summary failures

[eugene]

the stack summary analysis is failing to make the correct associations. for an example, see the stacks with labels of the form MD04.pspstest.2011021. This set of stacks (all with the same label, but 5 different filters) resulted in 4 stack summary entries. Of those, 3 generated jpegs (b1 & b2 versions) of a single skycell. The 4th generated a jpeg of the full set of skycells, but at least some of the skycells are from the wrong filter (some of the very ugly ones are from y-band, while the stack claims to be z-band). Eg, skycell.041 is missing in z-band (at the moment), but is included in the image from the y-band version (you can tell by the bad glint).

Finally, the ippMonitor image links are broken.