[FEA]: Limit pandas conversions in DFP modules/stages

[dagardner-nv]

Is this a new feature, an improvement, or a change to existing functionality?

Improvement

How would you describe the priority of this feature request

Low (would be nice)

Please provide a clear description of problem this feature solves

The DFP pipeline contains a few pandas conversions which can be costly in terms of performance, and requires storing a second copy of the cudf df.

Describe your ideal solution

Perform as many operations as possible in cudf.

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request