Unable to load KMeans model as original data type

**System Information (please complete the following information):**
- OS & Version: Windows 10
- ML.NET Version: ML.Net 1.7.0
- .NET Version: .Net 6.0

**Describe the bug**
After fitting a KMeans model, the trained model is of type TransformerChain>. After saving the model to a file using mlContext.Model.Save and reloading it using mlContext.Model.Load, there is no way to get the model back to TransformerChain>. The initial object loaded is of type TransformerChain. If you look at the object.LastTransformer, it is of type ClusteringPredictionTransformer>>. Casting either the loaded object or the LastTransformer gives a System.InvalidCastException. Using mlContext.Model.Load(...) as ... gives you null. Leaving LastTransformer as ITransformer does not provide you with the "Model" property.

**To Reproduce**

Code below for simple Console app, based on MS KMeans example.

```
using System;
using Microsoft.ML;
using Microsoft.ML.Data;
using Microsoft.ML.Trainers;
using System.Collections.Generic;
using System.IO;

namespace KMeansSample
{
internal class Program
{
static void Main(string[] args)
{
// Create a new context for ML.NET operations. It can be used for
// exception tracking and logging, as a catalog of available operations
// and as the source of randomness. Setting the seed to a fixed number
// in this example to make outputs deterministic.
var mlContext = new MLContext(seed: 0);

// Create a list of training data points.
var dataPoints = GenerateRandomDataPoints(1000, 0);

var _model = mlContext.Clustering.Trainers.KMeans(numberOfClusters: 3, featureColumnName: "Features");

// Convert the list of data points to an IDataView object, which is
// consumable by ML.NET API.
IDataView trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
var dataProcessPipeline = mlContext.Transforms.Concatenate("Features", nameof(DataPoint.val1),
nameof(DataPoint.val2),
nameof(DataPoint.val3),
nameof(DataPoint.val4)).AppendCacheCheckpoint(mlContext);
var trainingPipeline = dataProcessPipeline.Append(_model);

var trainedModel = trainingPipeline.Fit(trainingData);

mlContext.Model.Save(trainedModel, trainingData.Schema, "C:\\temp\\trainedModel.ms");

//This version will produce a System.InvalidCastException
using (var stream = new FileStream("C:\\temp\\trainedModel.ms", FileMode.Open, FileAccess.Read, FileShare.Read))
{
var loadedModel = (TransformerChain>)(mlContext.Model.Load(stream, out var modelInputSchema1));
var lastTransformer = loadedModel.LastTransformer;
var modelParameters = lastTransformer.Model;
}

//This version will produce a null reference exception
using (var stream = new FileStream("C:\\temp\\trainedModel.ms", FileMode.Open, FileAccess.Read, FileShare.Read))
{
var loadedModel = mlContext.Model.Load(stream, out var modelInputSchema1) as TransformerChain>;
var lastTransformer = loadedModel.LastTransformer;
var modelParameters = lastTransformer.Model;
}

//This version gives a compiler error

using (var stream = new FileStream("C:\\temp\\trainedModel.ms", FileMode.Open, FileAccess.Read, FileShare.Read))
{
var loadedModel = mlContext.Model.Load(stream, out var modelInputSchema1) as TransformerChain ;
var lastTransformer = loadedModel.LastTransformer;
var modelParameters = lastTransformer.Model;
}

//This version also gives a compiler error
using (var stream = new FileStream("C:\\temp\\trainedModel.ms", FileMode.Open, FileAccess.Read, FileShare.Read))
{
var loadedModel = mlContext.Model.Load(stream, out var modelInputSchema1);
var lastTransformer = loadedModel.LastTransformer;
var modelParameters = lastTransformer.Model;
}

}

private static IEnumerable GenerateRandomDataPoints(int count,
int seed = 0)
{
var random = new Random(seed);
float randomFloat() => (float)random.NextDouble();
for (int i = 0; i < count; i++)
{

yield return new DataPoint
{
val1 = randomFloat(),
val2 = randomFloat(),
val3 = randomFloat(),
val4 = randomFloat()

};
}
}

// Example with label and 50 feature values. A data set is a collection of
// such examples.
private class DataPoint
{

public float val1 { get; set; }
public float val2 { get; set; }
public float val3 { get; set; }
public float val4 { get; set; }

}

// Class used to capture predictions.
private class Prediction
{
// Original label (not used during training, just for comparison).
public uint Label { get; set; }
// Predicted label from the trainer.
public uint PredictedLabel { get; set; }
}

// Pretty-print of ClusteringMetrics object.
private static void PrintMetrics(ClusteringMetrics metrics)
{
Console.WriteLine($"Normalized Mutual Information: " +
$"{metrics.NormalizedMutualInformation:F2}");

Console.WriteLine($"Average Distance: " +
$"{metrics.AverageDistance:F2}");

Console.WriteLine($"Davies Bouldin Index: " +
$"{metrics.DaviesBouldinIndex:F2}");
}
}
}

```

**Expected behavior**
Able to load object back into the same type as the object that was saved.