pub fn temp_big_float_zero( precision_bits: u64, stack: &mut PodStack, ) -> (&mut BigFloat, &mut PodStack)