Evaluator

[gabeorlanski]

Fixes #5405.

Changes proposed in this pull request:

• New Evaluator Base class that serves as a way to evaluate and post-process a single data set.
  The new Postprocessor Base Class will post-process both batch inputs and output predictions into a human-readable string format.
• For both, SimpleEvaluator and SimplePostprocessor are implemented to replicate the original evaluation behavior.
• Integrated it with the Evaluate class.
  • Users will have a new top-level key in their configs called "evaluation" that maps directly to an Evaluator` sub-class. (Placeholder poping of this from the config was added to the train function so that it does not break).
  • For the time being, it is not required by evaluate to have this key and will default to the original behavior.
• Added better handling for multiple file evaluation so that users need only pass the input files and then use a CLI arg to automatically generate predictions and output files.
• Fixed the tests for multi-file evaluation to properly test multi-file evaluation.
• Probably a few others I am missing.

Before submitting

• I've read and followed all steps in the Making a pull request
  section of the CONTRIBUTING docs.
• I've updated or added any relevant docstrings following the syntax described in the
  Writing docstrings section of the CONTRIBUTING docs.
• If this PR fixes a bug, I've added a test that will fail without my fix.
• If this PR adds a new feature, I've added tests that sufficiently cover my new functionality.

After submitting

• All GitHub Actions jobs for my pull request have passed.
• codecov/patch reports high test coverage (at least 90%).
  You can find this under the "Actions" tab of the pull request once the other checks have finished.

[gabeorlanski]

@AkshitaB Sorry for the delay but here is the draft. Handling extra metrics is not yet implemented, but I felt like this was already big enough to warrant a PR

[AkshitaB]

@gabeorlanski Looks great, overall! I left some comments/suggestions.