Highlight orphan bodies containing T-bar(s) that have orphan partners

[takemuras]

After 0.5 proofreading, it would be useful to highlight orphan bodies with T-bar(s) in neuTu that have orphan postsynaptic partners - this might show where the reconstruction is bad - so that we could see if there is any region that is worse segmented than another.

Also nice, by calculating the percentage of identified PSDs per T-bar (how may PSDs are 0.5 bodies?), to differentiate T-bar orphans that have more impact to connectivity. For instance T-bar orphan with >50% PSDs are highlighted in red, the others are in yellow.