Make docs more clear that string is list of utf-16 code points

[richyliu]

This would close #1047

[showell]

I think the correct terminology is "UTF-16 code units", not "UTF-16 code points." Or you could say "Unicode code points encoded with UTF-16".

[showell]

To get this merged, it probably makes sense to address my most recent comment and then squash your two commits together.

I think we want to be really precise in this documentation, since there are subtle differences between Unicode code points and the encoding of Unicode characters. It's also possible that we want to underplay the UTF-16 aspect, since that seems like mostly an internal implementation detail.

@richyliu You may find this useful:

https://www.quora.com/In-the-Unicode-standard-what-is-the-difference-between-a-code-unit-and-a-code-point

I'm not super clear on Unicode terminology myself, but I know enough to spot some things that are ambiguous.

It's possible that Evan or somebody else in the core team will just want to revisit the docs holistically here.

[richyliu]

Ok, I changed line 84 back to code point and squashed the commits