Update syntax for Solidity 0.8.8

[haltman-at]

Syntax updates for Solidity 0.8.8:

1. I reclassified type as a keyword rather than a builtin, now that it can be used for custom type declarations.

2. I highlighed wrap and unwrap as builtins when used as a member of something else. I'm a little uncertain whether this is a good idea; it's possible that wrap and unwrap functions are sufficiently common that highlighting this would be annoying, and it would be better to just omit highlighting for these. So, uh, please let me know what you think of this. (@gnidan, I can't add you as a reviewer, but do you have an opinion on this?)

[cds-amal]

1. I reclassified type as a keyword rather than a builtin, now that it can be used for custom type declarations.

This seems reasonable.

2. I highlighed wrap and unwrap as builtins when used as a member of something else. I'm a little uncertain whether this is a good idea; it's possible that wrap and unwrap functions are sufficiently common that highlighting this would be annoying, and it would be better to just omit highlighting for these. So, uh, please let me know what you think of this. (@gnidan, I can't add you as a reviewer, but do you have an opinion on this?)

I think if they are sufficiently common it is a good thing to draw attention to, especially as it is a new feature.

[cds-amal]

Looks good to me!

[cds-amal]

This was pre-existing but it would be nice to have comments that explain where these groupings come from: Address / user types, type information etc

[haltman-at]

Sure, I could add that!

[haltman-at]

I think if they are sufficiently common it is a good thing to draw attention to, especially as it is a new feature.

I'm not sure I understand what you're saying here. I'm saying that if x.wrap and x.unwrap are sufficiently common as existing things, it might be obnoxious to highlight. Ideally not every instance of x.wrap or x.unwrap would be highlighted, only those cases where x is a UDVT. But that's not something we can feasibly do with HLJS; it's either highlight all such cases or none of them. So the question is, would making it highlighted in all cases -- to get it highlighted when x is a UDVT -- be too obnoxious due to existing (and non-conflicting) usage of x.wrap or x.unwrap in other cases?

I don't think I understand what your answer to that question is.

[cds-amal]

But that's not something we can feasibly do with HLJS; it's either highlight all such cases or none of them.

Oh, right! There's not enough information to disambiguate. I misunderstood your 2nd point. However I still think that if this is a new feature we should draw attention to it and wait for feedback that it is noisy.

This Github search suggests wrap is not so popular in GH public repos

[haltman-at]

OK, cool, I'll go ahead with that then. I'll go back and add the comments you suggested and then merge / release this.

I asked about wrap/unwrap because earlier I had contemplated making value a builtin (#24) and @frangio was like no that'd be way too noisy. (We ended up coming up with a special-case hack to handle it.) I knew that like wrapped ether was a thing so I wasn't sure if that would be a common existing usage that would cause noise? Wrap/unwrap also just seemed like it might be a common pair in general. But sounds like it isn't so there's no problem!