New node-red control flow for Kubeflow customization pipeline execution

We develop a new Node-red pipeline control flow over kubeflow pipeline execution. This flow divides the AI training, testing and deployment meta format for kubeflow AI pipeline execution. The functionality of each node varies depending on its assigned mission. For example, the data read node allows loading of different training data sets. The pre-process node allows data cleaning, regulation, normalization options. Further, the algorithm node allows selection of different AI training algorithms, such as decision tree, random forest, logistic regression etc. Details and the code is included in the following repo. site.
https://github.com/sefgsefg/Federated-Learning-on-kubeflow-with-nodered/tree/main/FL_kubeflow_with_node-red/examples

Start by reviewing the examples in the linked Federated-Learning-on-kubeflow-with-nodered repository, since no files or tests are identified in this issue. Determine the required Node-RED control flow and node responsibilities for Kubeflow training, testing, and deployment; done means a complete working flow covering those stages.

Written by the indexing model from the issue text.