Constructs a FixedPoint number from another number, or a string.
If value is a string, then the string is parsed and the precision is taken from the number of digits after the decimal point, string, ignoring decimals.
If value is a floating point, then the result is rounded to the nearest fixed-point value, given the precision indicated in decimals, which defaults to 2.
If value is an int, the decimals defaults to 2.
If value is a FixedPoint, the precision of the result will be the max of decimals or the precision of value.