Choose type for integer literals to match the target mask size (if possible).

On a target with a 16-bit mask (for example), we would choose the type
of an integer literal "1024" to be an int16.  Previously, we used an int32,
which is a worse fit and leads to less efficient code than an int16
on a 16-bit mask target.  (However, we'd still give an integer literal
1000000 the type int32, even in a 16-bit target.)

Updated the tests to still pass with 8 and 16-bit targets, given this
change.

tests/kilo-mega-giga-2.ispc

@@ -8,5 +8,5 @@ export void f_fu(uniform float RET[], uniform float aFOO[], uniform float b) {
 }
 
 export void result(uniform float RET[]) {
-    RET[programIndex] = 2*1024*1024 + 5;
+    RET[programIndex] = 2ul*1024ul*1024ul + 5;
 }