Scanner and parser use different tables for constants

Moreover, each function being parsed has its own table.

The code is cleaner when each table is used for one specific purpose:
The scanner uses its table to anchor and unify strings, mapping strings
to themselves; the parser uses it to reuse constants in the code,
mapping constants to their indices in the constant table. A different
table for each task avoids false collisions.

lparser.h

@@ -146,6 +146,7 @@ typedef struct FuncState {
   struct FuncState *prev;  /* enclosing function */
   struct LexState *ls;  /* lexical state */
   struct BlockCnt *bl;  /* chain of current blocks */
+  Table *kcache;  /* cache for reusing constants */
   int pc;  /* next position to code (equivalent to 'ncode') */
   int lasttarget;   /* 'label' of last 'jump label' */
   int previousline;  /* last line that was saved in 'lineinfo' */