Zero-copy string deserialization

Over on Reddit, @mikeyhew mentioned that there might be an option to parse JSON strings without copying:

Just wanted to point out that serde-json isn't zero-copy because it will copy strings to turn escape sequences like "\n" and "" into the character they represent. To parse JSON without copying, you could make a custom string type, JsonStr, which is utf-8 like str but can contain escape sequences.

I forgot about that, but it's actually a great idea! Here's the upstream discussion on serde-json. We should give this custom string type some serious consideration, as string allocation takes a big part of the encoding/decoding process at the moment.

If anyone wants to give it a shot, go for it.