Redefining int_t and real_t

Integer is defined as int, and real is defined as float by default in MY-BASIC, which fits most situations. Besides it's possible to redefine them with other precision as you need.

There are some data precision related macros in my_basic.h need to be modified, as follow:

#ifndef int_t
#	define int_t int
#endif
#ifndef real_t
#	ifdef MB_DOUBLE_FLOAT
#		define real_t double
#	else /* MB_DOUBLE_FLOAT */
#		define real_t float
#	endif /* MB_DOUBLE_FLOAT */
#endif /* real_t */

#ifndef mb_strtol
#	define mb_strtol(__s, __e, __r) strtol((__s), (__e), (__r))
#endif
#ifndef mb_strtod
#	define mb_strtod(__s, __e) strtod((__s), (__e))
#endif

#ifndef MB_INT_FMT
#	define MB_INT_FMT "%d"
#endif
#ifndef MB_REAL_FMT
#	define MB_REAL_FMT "%g"
#endif

#ifndef MB_FNAN
#	ifdef MB_DOUBLE_FLOAT
#		define MB_FNAN 0x7ff8000000000000
#	else /* MB_DOUBLE_FLOAT */
#		define MB_FNAN 0x7fc00000
#	endif /* MB_DOUBLE_FLOAT */
#endif /* MB_FNAN */
#ifndef MB_FINF
#	ifdef MB_DOUBLE_FLOAT
#		define MB_FINF 0x7ff0000000000000
#	else /* MB_DOUBLE_FLOAT */
#		define MB_FINF 0x7f800000
#	endif /* MB_DOUBLE_FLOAT */
#endif /* MB_FINF */

MY-BASIC uses int_t and real_t for all internal numeric declarations, therefore changing the definition of them will affect all numeric storage and evaluation. mb_strtol and mb_strtod are used to parse numbers. MB_INT_FMT and MB_REAL_FMT are used to output int_t and real_t in properly ways, you need to redefine them according to specific precision, don't forget to redefine MB_FNAN and MB_FINF for NaN and INF real values as well.

Added an MB_DOUBLE_FLOAT macro to simply redefine real_t as double precision.