Confusable checking for CJK Ideographs

[crlf0710]

Currently the confusable data doesn't fully include CJK ideographs related info.
Unihan database provides these:

• kZVariant which describes logically unifiable ideographs that are separately encoded for other reasons
• kSpoofingVariant which describes confusable ideograph relationships
• EquivalentUnifiedIdeograph.txt which describes CJK Radical/Stroke and their corresponding Ideograph.

Procedurally i'd recommend Unicode security standards adopt these as confusables upstream, either by data or by algorithm, and then this crate can support them automatically.

cc @Manishearth

[Manishearth]

Neat idea!

I don't know what it would take for Unicode to do this, and whether it wants to. I think Han confusables are a bit different in a couple ways (a lot of Han confusables show up in regular text, even side by side!) so that might be why they are separate. We could still parse unihan and support it though, unihan isn't too complicated.