[trace-obfuscation] Add SQL obfuscator and make other obfuscators run faster

[paullegranddc]

What does this PR do?

This PR:

• Adds a simple SQL obfuscator based on the java and dotnet implementations
• Refactor the existing redis obfuscator to use less allocations, bringing a nice (60%) gain in performance
• Refactor the existing trace tags obfuscator to use less allocations and pre-compute some values, bringing a nice (40%) gain in performance
• Refactor the trace-obfuscation benches and adding some new ones.

The changes to the existing redis and tags obfuscators should not affect their semantics as they are still passing the existing tests.

Motivation

An SQL obfuscator is needed to make the stats computation in data-pipeline implementation spec compliant.

Additional Notes

Anything else we should know when reviewing?

How to test the change?

Describe here in detail how the change can be validated.

For Reviewers

• If this PR touches code that signs or publishes builds or packages, or handles credentials of any kind, I've requested a review from @DataDog/security-design-and-guidance.
• [ x ] This PR doesn't touch any of that.

[bantonsson]

Nice work. My only real comment is about the wiring up and running of the benchmarks, standard documented rust way is not working anymore.