Restore scheduler functionality

[marrrcin]

Scheduler was temporarily disabled due to slight issues with MLFlow integrations (token obtaining) and not working re-scheduling (old schedules were not disabled).
Restore this functionality in future releases.

[Lasica]

Is this still relevant? I don't understand the context of this issue.

[Stormz4]

Are the any news about this issue?

[marrrcin]

We could potentially use https://cloud.google.com/vertex-ai/docs/pipelines/schedule-pipeline-run#aiplatform_create_pipeline_schedule_sample-python_vertex_ai_sdk 🤔

Let me take it to the team and we'll get back to you :)

[Stormz4]

I solved this problem with the following workaround:

1. Compile the Kedro project with kedro vertexai compile
2. Then, through a python script:
   1. Create a PipelineJob object
      (https://cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform.PipelineJob)
   2. Call the function create_schedule of said PipelineJob (https://cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform.PipelineJob#google_cloud_aiplatform_PipelineJob_create_schedule)

It should be pretty streamlined to implement said solution in Kedro's API.

optional step suggestion (could be a flag inserted through args):

• Check if a job already exists with my desired display name: if it exists, delete it (and recreate it). This is needed since in vertexai you can have multiple pipeline with the same name