You can't coerce a float to an integer, because they want to you decide
whether you want to use round, floor, ceiling, and so forth. Sometimes, though,
you know or believe that the float is the floating point representation of an
integer, so all of the above would produce the same result. This function is
for that case. It coerces `number' to an integer, even if it's a float. The
fractional part of the float must be zero or an error is signalled.