Simplify user experience of AML pipeline tuning.

The approach of Azure pipeline tuning described in https://microsoft.github.io/FLAML/docs/Examples/Tune-AzureML-pipeline/ needs the users to set up the tuning pipeline.

The issue is targeted to minimize the customization requirements and make the tuning component as generic as possible.

The list of to-be-improves include:
- Make the approach applicable for both whole pipeline tuning and subgraph tuning and step tuning.