[Patch 3/3] Separate `repeating` node from `alternative` node in regex

[Jonathan Wakely]

On 25/04/14 18:04 -0400, Tim Shen wrote:
>    	* include/bits/regex_automaton.h (_NFA<>::_M_insert_repeat):
>    	Add _S_opcode_repeat support to distingush a loop from
>    	_S_opcode_alternative.
>    	* include/bits/regex_automaton.tcc (_State_base::_M_print,
>    	_State_base::_M_dot, _NFA<>::_M_eliminate_dummy,
>    	_StateSeq<>::_M_clone): Likewise.
>    	* include/bits/regex_compiler.tcc (_Compiler<>::_M_quantifier):
>    	Likewise.
>    	* include/bits/regex_executor.tcc (_Executor<>::_M_dfs): Likewise.
>    	* include/bits/regex_scanner.tcc (_Scanner<>::_M_eat_escape_ecma):
>    	Uglify local variable __i.
>    	* include/bits/regex_compiler.h (_BracketMatcher<>::_M_make_cache):
>    	Use size_t instead of int to compare with vector::size().

This one is OK to commit too - thanks.