Standardise column names

[mmc46]

neuprint_search() returns columns

 [1] "bodyid"        "name"          "type"          "status"        "voxels"       
 [6] "pre"           "post"          "cropped"       "soma"          "cellBodyFiber"

whereas neuprint_find_neurons() returns

[1] "bodyid"       "bodyname"     "bodytype"     "neuronStatus" "size"         "npre"        
[7] "npost"        "CA(R).pre"    "CA(R).post"  

The latter query was for neurons with inputs in CA(R).

ca.info = neuprint_find_neurons(
  input_ROIs = "CA(R)",
  output_ROIs = NULL,
  all_segments = FALSE # if true, fragments smaller than 'neurons' are returned as well
)

[jefferis]

@romainFr @alexanderbates I think we should be moving to a new release of the package after which we reduce the rate of churn. Do we think it would be a good idea to implement this change before that?

[romainFr]

Yes. I'd be in favor of the neuprint_search convention. Also it would be good to return the same kind of format for the ROI pre and post as neuprint_connection_table (meaning having roi and ROIweight columns instead of the columns named with ROI.pre)?