Optimize String.similarity by avoiding allocation for bigrams.

[Ivorforce]

The logic is the same; there's just a lot less copying than before.

Relevant similarity benchmark goes down from 530 to 80 (6.6x increase in speed).

I also tested longer strings. In this example, the benchmark goes down from 48019us to 2417us (20x increase in speed):

	String s1 = "Lorem ipsum anyone?";
	String s2 = "Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium doloremque laudantium, totam rem aperiam, eaque ipsa quae ab illo inventore veritatis et quasi architecto beatae vitae dicta sunt explicabo. Nemo enim ipsam voluptatem quia voluptas sit aspernatur aut odit aut fugit, sed quia consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. Neque porro quisquam est, qui dolorem ipsum quia dolor sit amet, consectetur, adipisci velit, sed quia non numquam eius modi tempora incidunt ut labore et dolore magnam aliquam quaerat voluptatem. Ut enim ad minima veniam, quis nostrum exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid ex ea commodi consequatur?";

	auto t0 = std::chrono::high_resolution_clock::now();
	for (int i = 0; i < 1000; i ++) {
		// Test
		int x = s1.similarity(s2);
	}
	auto t1 = std::chrono::high_resolution_clock::now();
	std::cout << std::chrono::duration_cast<std::chrono::microseconds>(t1 - t0).count() << "us\n";

[aaronp64]

Looks good to me. Speed difference makes sense, and did some basic tests to confirm same results from old and new code:

func test_similarity():
	print("test1".similarity("test2"))
	print("test12".similarity("test23"))
	print("abc12".similarity("def12"))
	print("12abc".similarity("12def"))
	print("abc".similarity("def"))

[kiroxas]

Definitely better than all the allocations in bigrams !

[clayjohn]

Faster, easier to read, avoids allocations. All very nice :)

[Repiteo]

Thanks!