Parsing of IPv6 addresses seems to incorrectly include `[]`s

• Note that this crate implements the URL Standard not RFC 1738 or RFC 3986

Describe the bug

Hi all!

It seems like parsing a URL that contains an IPv6 address like http://[::1] returns a Url struct that, when .host().unwrap.to_string() is called on it, returns [::1] where the docs state, for Host:

An IPv6 address. Url::host_str returns the serialization of that address between [ and ] brackets, in the format per RFC 5952 A Recommendation for IPv6 Address Text Representation: lowercase hexadecimal with maximal :: compression.

The "between" language here makes me think that it should not be including the [].

According to the linked URL Standard, it also states, for host parsing, that:

If input starts with U+005B ([), then:

If input does not end with U+005D (]), validation error, return failure.

Return the result of IPv6 parsing input with its leading U+005B ([) and trailing U+005D (]) removed.

Which makes it seem like the [] should be removed.

I'm just curious to get more clarity on expected behavior here.

This came up due to some confusion in https://github.com/nats-io/nats.rs/pull/386

Start with the Url::host(), Url::host_str(), and Host::to_string() behavior for the input http://[::1], then compare it with the linked URL Standard and Host documentation. Add or update a regression test for the IPv6 host representation; done means the implementation, documentation, and test agree on whether brackets are included.