Fix: Estimator.logs()

If you call any `Estimator.logs()` either `Pytorch` you will receive the following error

```python
ParamValidationError: Parameter validation failed:
Invalid type for parameter TrainingJobName, value: <sagemaker.estimator._TrainingJob object at 0x7ffc4af0d520>, type: <class 'sagemaker.estimator._TrainingJob'>, valid types: <class 'str'>
```

adjusted it to `self.latest_training_job.name` and it works

Author: philschmid

src/sagemaker/estimator.py

@@ -815,7 +815,7 @@ def logs(self):
         If the output is a tty or a Jupyter cell, it will be color-coded based
         on which instance the log entry is from.
         """
-        self.sagemaker_session.logs_for_job(self.latest_training_job, wait=True)
+        self.sagemaker_session.logs_for_job(self.latest_training_job.name, wait=True)
 
     def deploy(
         self,