Benchmarks needed

As I mentioned in #109, a benchmark suite is important for ensuring a stable level of performance.

Personally, I've been using https://github.com/google/swift-benchmark - it's a little rough around the edges, but it produces decent reports on all platforms and the dynamic iterations features works really well for getting stable readings out of small code snippets. I'd encourage Apple (or Swift.org) to adopt it, given S4TF's demise, the likelihood that the original authors will abandon it, and the need for good benchmarking utilities that work across platforms.

Start by reading issue #109 and reviewing the google/swift-benchmark project linked here, including its cross-platform reporting and dynamic-iteration behavior. Define the benchmark suite's scope and how it should fit swift-algorithms; done means the project has stable performance measurements across platforms.

Written by the indexing model from the issue text.