How can I prevent JRS report fields with null values from creating inconsistent xml structure for Excel connection

I have noticed that when a field may or may not be null (due to either an unpopulated custom attribute in RTC or an optional join with no data) that the report structure (xml) will change depending whether or not there is actually a null value in the returned data.  This creates a problem using the data in Excel through Power Query, since changing the structure essentially breaks the query.   I have been able to work around this by using a custom expression such as COALESCE instead of using the attribute directly.  However, I now have a report that can have a null value due to either the attribute OR the optional join, and the COALESCE statement works only for the optional join null, not the missing attribute value null - those still come back null.  I have tried nesting COALECE and various combinations of NVL, etc, and I haven't found anything that works.  How can I address this?

Using JRS:

Version: 6.0.4

Build ID: JRS_6.0.4-I20180129-1426