Case-insensitive grammars support

[Eliah-Lakhin]

Lady Deirdre's regex syntax currently does not have a straightforward way to express case-insensitive keywords in the Token macro.

To address this issue, a new operator i(...) will be introduced. This operator will turn every encountered Unicode (and ASCII in particular) character inside the surrounding expression into a case-insensitive choice if the character has a case property.

For example, i("abc") will be semantically equal to ['a', 'A'] & ['b', 'B'] & ['c', 'C'].

(See prior discussion in #17)