calc: C-style arbitrary precision system Files

The following are the changes in this release:

    Ported calc to the s390x IBM Mainframe running RHEL9.1.

    Added cal/splitbits.cal:

	splitbits(x, b)

	Given an integer x, split the value into a list of integers,
	each of which is at most b bits long.

	The value b must be an integer > 0.

	The number of elements in the returned list is:

	    ceil((highbit(x) + 1) / b)

	If x == 0, then a list of 1 element containing 0 is returned.

	If x < 0, then the two's compliment of abs(x) is returned.
	Even though calc represents negative integers as positive values
	with sign bit, the bits returned by this function are as if
	the integer converted as if the integer was a two's compliment value.

	See also the help command:

	    ; help resource