Remove NaNs from any resulting fit, regardless of the source to ensur…

…e resulting data is plottable

src/sas/qtgui/Perspectives/Fitting/FittingWidget.py

@@ -83,6 +83,7 @@ def get_weights(self: Any, name: str) -> Tuple[np.ndarray, np.ndarray]:
 
 logger = logging.getLogger(__name__)
 
+
 class ToolTippedItemModel(QtGui.QStandardItemModel):
     """
     Subclass from QStandardItemModel to allow displaying tooltips in
@@ -3116,6 +3117,12 @@ def createTheoryIndex(self, fitted_data):
         Create a QStandardModelIndex containing model data
         """
         name = self.nameFromData(fitted_data)
+        # TODO: Temporary Hack to fix NaNs in generated theory data
+        #  This is usually from GSC models that are calculated outside the Q range they were created for
+        #  The 'remove_nans_in_data' should become its own function in a data utility class, post-6.0.0 release.
+        from sasdata.dataloader.filereader import FileReader
+        temp_reader = FileReader()
+        fitted_data = temp_reader._remove_nans_in_data(fitted_data)
         # Modify the item or add it if new
         theory_item = GuiUtils.createModelItemWithPlot(fitted_data, name=name)
         self.communicate.updateTheoryFromPerspectiveSignal.emit(theory_item)