With the new equal operator in Arithmetic, we can choose certain pixels in the image, except blank pixels (NaN in float types and fixed constants for other types).

The problem with blank values is that in floating points, the IEEE NaN standard defines this value as failing any conditional operator, so NaN is not equal to its self and the only reliable way we can check for it is the `isnan' function. For integer types the blank value is a constant (defined by the type in Gnuastro's `lib/gnuastro/fits.h'. But we don't want the users to have to check that value manually.

One solution to these problems might be this: we add a unique type of operand (not a file name or number) called `blank' (the name can be changed). Then when the equal operator confronts this operand (either as the first popped operand or the second), it will use `isnan', not the equal operator. Internally (for the time being, see task #13869), in Arithmetic, everything is in double type, so the single `isnan' is enough. Later it can be expanded to accommodate all types.