Add some metadata to analysis statistics to identify which analyses are outputting to PI Points, which are configured as an Analysis Data Reference and which are a mix of both. Currently this isn't identified and can cause some confusion. For example, an analysis DR could be reported as skipping, but if it isn't used as the input of another analysis this would not necessarily cause any issues. It may be expensive for an AF Client to retrieve the data. It's however not accurate to say that the Analysis DR is skipping.
