Component that walks through the page tree, finds resource dictionaries and removes any image xobject (type xobject, subtype image) that is not wrapped by a
ReadOnlyFilteredCOSStream
and any font that is not wrapped by a
InUseFontDictionary
. This is the step performed after pages content stream have been already
parsed, all used image wrapped by a
ReadOnlyFilteredCOSStream
and all used fonts wrapped by a
InUseFontDictionary
and placed back to the resource dictionary.
This is done in two steps because dictionaries can be shared/inherited by pages so we can't take a single page, identify used images and remove the remaining because that same
resource dictionary can be used by other pages.