Scrubbing learners (Meta issue)

[Ivanidzo4ka]

We need to clean our API for learners.
This includes following things:

1. No protected fields/members/method in public classes. Only private protected.

2. Learner class is sealed.

3. The trainer name types should follow the names used in the contexts (see The trainer name types should follow the names used in the contexts #2172)

4. ModelPameters for this Learner is also clean. (sealed, public documentation, no public constructor)

5. Options cleaning:

• They should be named Options, and all their base classes (except LearnerInputBase*)
• Option should have meaning and proper way it initialize it self. We shouldn't accept int if in reality we use enum (FastTree EarlyStoppingMetric is an int but only accepts specific values #2521) or accept array of ints as a string separated by comma public string CustomGains = "0,3,7,15,31";
• No short names.
• Standard names like:
  /// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
NumberOfIterations, NumberOfThreads, LearningRate,L2Regularization, L1Regularization
  no (MaxIterations,NumIterations, NumThreads, L2Weight, L1Weight)

6. If we communicate with user (exceptions, channel messages), don't use DataKind values. No R4, U4, and so on, it should be float, uint, etc.

ignore for now:

1. Everything public has proper documentation.

2. We have samples of how use base call to trainer (no options) and call with options.

3. We have baseline tests for learner.

https://docs.microsoft.com/en-us/dotnet/standard/design-guidelines/general-naming-conventions#using-abbreviations-and-acronyms

`

[rogancarr]

9. There should be consistency within Tasks (e.g. BinaryClassification, Regression) on how labels are specified (e.g. R4, bool, KeyType).

For instance, many learners accept R4 values and helpfully convert them to bool or KeyType, while some learners only accept KeyType. (see #2628 for a multiclass example). This inconsistency is confusing from a user perspective because you can't hot-swap Trainers in a pipeline.

[wschin]

We also have rowGroupColumnName to replace groupIdColumn and exampleWeightColumnName for weightColumn.

[rogancarr]

The classnames themselves come in many different styles and they are surfaced through options:

For example, SDCA and FastTree have very different class names for binary classification:

var sdca = mlContext.BinaryClassification.Trainers.StochasticDualCoordinateAscent(
    new SdcaBinaryTrainer.Options { NumberOfThreads = 1 });

var fastTree = mlContext.BinaryClassification.Trainers.FastTree(
    new FastTreeBinaryClassificationTrainer.Options { NumberOfThreads = 1 });

These class names should have a standard grammar.