Range gate optimization

[vezenovm]

Problem

After this PR in barretenberg is merged (AztecProtocol/barretenberg#369) an inefficiency has been introduced. Whenever there is an overlapping range check on the same witness we have to add an arithmetic gate.

Proposed solution

We can avoid this additional arithmetic gate by optimizing out redundant range constraints. We can remove the less restrictive range checks and keep the most restrictive. For example, if we range check the value at witness index 90 with a bit size of 4 and 8 in two different constraints we can remove the range constraint with bit size of 8.

Alternatives considered

No response

Additional context

No response

Submission Checklist

• Once I hit submit, I will assign this issue to the Project Board with the appropriate tags.