[Feature Request]: User can specify the project where BigQueryIO.fromQuery() will be executed

### What would you like to happen?

By default the query is executed in the same project where the job is running. Often a different project (associated with a particular BigQuery reservation) should be used.

The "bigQueryProject" pipeline option can be used to change the execution project. But it would be better to explicitly specify it on BigQueryIO connector, e.g., BigQueryIO.readTableRows().from("...").usingExecutionProject("..."). This will allow per transform configuration and should prevent accidental execution of query in the wrong project because the "bigQueryProject" option applies to all transforms and used for both reads and writes.

### Issue Priority

Priority: 3 (nice-to-have improvement)

### Issue Components

- [ ] Component: Python SDK
- [X] 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