[Bug]: SyntheticUnboundedSource missing records

[akashorabek]

What happened?

It's a bug in the implemenation of SyntheticUnboundedSource class. By default, it splits the data to some number of bundles and lost records number in write streaming pipeline equals to that number of bundles.

Various stress and load tests use this Source, for example, KafkaIOLT and PubSubIOLT. To reproduce the bug, you can run PubSubIOLT (for convenience, I recommend running it on DirectRunner) and check the value of numRecords. It is typically lower than what we initially set.

Issue Priority

Priority: 3 (minor)

Issue Components

• [ ] Component: Python SDK
• [X] Component: Java SDK
• [ ] Component: Go SDK
• [ ] Component: Typescript SDK
• [ ] Component: IO connector
• [ ] Component: Beam YAML
• [ ] 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