integer types use

[vladak]

Add a module on when to use size_t and when other integer types.

E.g. https://stackoverflow.com/questions/131803/unsigned-int-vs-size-t

Plus examples on working with file offset (seek, read, write, ...), etc.
The size_t type is the unsigned integer type that is the result of the sizeof
operator (and the offsetof operator), so it is guaranteed to be big enough to
contain the size of the biggest object your system can handle (e.g., a static
array of 8Gb).

Also add an idea to use 'u' for an unsigned integer in for

[vladak]

Related to the int/size_t discussion above, make a new module for uint32_t et
al. It is in the spec as part of stdint.h.

7.18.1.1 Exact-widthinteger types1The typedef nameintN_tdesignates a signed
integer type with widthN,nopaddingbits, and a two’scomplement
representation.Thus,int8_tdenotes a signed integertype with a width of
exactly 8 bits.2The typedef nameuintN_tdesignates an unsigned integer type
with widthN.Thus,uint24_tdenotes an unsigned integer type with a width of
exactly 24 bits.3These types are optional.However, ifanimplementation
provides integer types withwidths of 8, 16, 32, or 64 bits, no
padding bits, and (for the signed types) that have atwo’scomplement
representation, it shall define the corresponding typedef names