the current code assumes that float.Parse behaves the same as the
internal C++ code, however without using "end" as part of the span, it
will parse from index 0 to the end of the string, ignoring commas. for
example, this causes it to parse "0,5,0" with divisor "," as [50, 50,
0], as the float.Parse method ignores commas in floats.
if another divisor is used, it throws a System.FormatException due to
containing invalid characters, as it fails to account for the position
of the divisor for the span.
b5591bf5f6