tdf#76649 Skip NaN initial values for min and max

min and max were initiated as aValuesX[0] which could be NaN
test i before evaluating aValuesX[i]

Change-Id: I0246913dcf427e59fd354321eeffa7c08c41c08d
Reviewed-on: https://gerrit.libreoffice.org/17772Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

chart2/source/view/main/VDataSeries.cxx

@@ -520,9 +520,13 @@ void VDataSeries::getMinMaxXValue(double& fMin, double& fMax) const
 
     if(aValuesX.getLength() > 0)
     {
-        fMax = fMin = aValuesX[0];
+        sal_Int32 i = 0;
+        while ( i < aValuesX.getLength() && ::rtl::math::isNan( aValuesX[i] ) )
+            i++;
+        if ( i < aValuesX.getLength() )
+            fMax = fMin = aValuesX[i++];
 
-        for (sal_Int32 i = 1; i < aValuesX.getLength(); i++)
+        for ( ; i < aValuesX.getLength(); i++)
         {
             const double aValue = aValuesX[i];
             if ( aValue > fMax)