Add Bytes32x2Set

[Amxx]

• Refactor EnumerableSet.opt.js to support arrays of basetype
  • doesn't affect currently generated code (backward compatible)
• Use the refactor to implement Bytes32x2Set

PR Checklist

• Tests
• Documentation
• Changeset entry (run npx changeset add)

[changeset-bot]

🦋 Changeset detected

Latest commit: eb050b8

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
openzeppelin-solidity	Minor

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[arr00]

Suggested change

	mapping(bytes32 valueHash => uint256) _positions;
	mapping(bytes32 valueHash => uint256) _positionsPlusOne;

Thoughts on this for code clarity?

[Amxx]

I looked into the history of this piece of code:

1. EnumerableSet was introduced in 2.5. In that version, the code only supported addresses, and the structure was defined as:

    struct AddressSet {
        // Position of the value in the `values` array, plus 1 because index 0
        // means a value is not in the set.
        mapping (address => uint256) index;
        address[] values;
    }

2. In version 3.0, the inner/outer structure was used, with the default being a bytes32 set that is

    struct Set {
        // Storage of set values
        bytes32[] _values;

        // Position of the value in the `values` array, plus 1 because index 0
        // means a value is not in the set.
        mapping (bytes32 => uint256) _indexes;
    }

3. In version 5.0, the naming was changed to was we have currently (this is the PR). Note that the PR was the result of an audit issue.

4. This PR is just reusing the wording that we have in the other sets.

---

IMO:

• No user should touch that storage (if we could make it private we would)
• The description is really clear
• Overall, I tend to favor short names to long ones. IMO _positionsPlusOne is difficult/long to read and offer little benefit.
• Changing the name in the existing set would be breaking, and using a different name here would be inconsistent.

So I'd personally keep it that way. @ernestognw WDYT ?