Wonder if this is bug with the daylight saving hour - when a time period start from Oct ends in April was supply to the dl_saving input for sunpath component, it does not work as expected. Seems having no impact on the hoys results for a Dec 21st with and without the dl_saving input.
I just pushed a fix to the core libraries, which will ensure that Daylight Savings AnalysisPeriods with end dates earlier in the year from the start date are interpreted correctly:
You should be able to get the fix on your end with the LB Versioner in an hour or so.