feat: remove folding

[tserg]

What I did

How I did it

How to verify it

Commit message

Commit message for the final, squashed PR. (Optional, but reviewers will appreciate it! Please see our commit message style guide for what we would ideally like to see in a commit message.)

Description for the changelog

Cute Animal Picture

[codecov-commenter]

Codecov Report

Attention: 38 lines in your changes are missing coverage. Please review.

Comparison is base (bf26f83) 84.05% compared to head (fe925d6) 84.12%.

Files	Patch %	Lines
vyper/builtins/functions.py	80.29%	17 Missing and 10 partials ⚠️
vyper/ast/nodes.py	89.74%	2 Missing and 6 partials ⚠️
vyper/ast/folding.py	84.21%	2 Missing and 1 partial ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3709      +/-   ##
==========================================
+ Coverage   84.05%   84.12%   +0.07%     
==========================================
  Files          92       93       +1     
  Lines       13045    13160     +115     
  Branches     2925     2943      +18     
==========================================
+ Hits        10965    11071     +106     
- Misses       1654     1656       +2     
- Partials      426      433       +7     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[tserg]

I ran into a few complications:

• For literal nodes, we do not want the "folded_value" in metadata to be populated because this causes unnecessary duplication of some machinery like type-checking and type annotation. This requires the get_folded_value_maybe() and get_folded_value_throwing() to be tweaked for these nodes so that these functions return a valid node (i.e. they are literals and can be "folded") without populating the metadata.
• The consequence is that outside of the pre-typechecking phase, we should read the folded value using node._metadata.get("folded_value") because we are now interested in whether the node should actually be replaced by another folded node, and not whether it is capable of being folded.