doc: use "code", not "number", for token (and symbol) kinds

"Number" is too much about arithmethics.  "Code" conveys better the
"enum" nature of token kinds.  And of symbol kinds.

* doc/bison.texi: Here.

TODO

@@ -4,12 +4,8 @@
 - YYERRCODE, YYUNDEF, YYEOF
 - i18n in Java
 - symbol.type_get should be kind_get, and it's not documented.
-- token code instead of token user number etc.
 - YYERRCODE and "end of file" and translation
 
-** User token number, internal symbol number, external token number, etc.
-We could use "number" and "code".
-
 *** The documentation
 You can explicitly specify the numeric code for a token type...
 
@@ -18,9 +14,6 @@ The token numbered as 0.
 Therefore each time the scanner returns an (external) token number,
 it must be mapped to the (internal) symbol number.
 
-*** M4
-Make it consistent with the rest (it uses "user_number" and "number").
-
 ** Java: EOF
 We should be able to redefine EOF like we do in C.
 
@@ -57,6 +50,8 @@ exists.
 
 Rename endtoken as eoftoken.
 
+Also do it in data/skeletons.
+
 Don't rename in Bison 3.6 (it would be logical to do so) because that
 would probably create many conflicts in Vincent's work (see previous point).