[Task]: Remove archaic PubSubSource and PubSubSink

[robertwb]

What needs to happen?

Reading from and writing to PubSub should be proper (if primitive) transforms, rather than being represented by "special" parameters of Read/Write operations necessitating workarounds like https://github.com/apache/beam/blob/release-2.48.0/sdks/python/apache_beam/io/iobase.py#L946.

Issue Priority

Priority: 3 (nice-to-have improvement)

Issue Components

• [X] Component: Python SDK
• [ ] Component: Java SDK
• [ ] Component: Go SDK
• [ ] Component: Typescript SDK
• [X] Component: IO connector
• [ ] Component: Beam examples
• [ ] Component: Beam playground
• [ ] Component: Beam katas
• [ ] Component: Website
• [ ] Component: Spark Runner
• [ ] Component: Flink Runner
• [ ] Component: Samza Runner
• [ ] Component: Twister2 Runner
• [ ] Component: Hazelcast Jet Runner
• [ ] Component: Google Cloud Dataflow Runner

[Abacn]

Does this mean currently Python PubsubIO can only run on Dataflow runner (in addition to Python direct runner as there is a local overwrite) ?

[robertwb]

Correct, currently they only run on Dataflow (that swaps out PubSub) and the olde Python direct runner (that also swaps them out).