There were several limit names in the limit view builder that
appear to have been copied over from the split from nova. Each
name gets checked against the actual returned Cinder limit data
so this was adding a very small but unnecessary extra processing
while parsing the limits.
Pared names to check down to only those names actually being used
by Cinder.
Change-Id: I712f5dd047a533604645e655a80e542a05fb10e0