Add fake streaming source

[wengh]

Implement streamReader for FakeDataSource.

This pull request introduces significant enhancements to the FakeDataSource in the pyspark_datasources module, including the addition of a streaming capability and refactoring for better code organization. The most important changes include the addition of a streaming reader, the creation of a schema validation function, and updates to the documentation and tests.

Enhancements to FakeDataSource:

• Added a streaming reader with the FakeDataSourceStreamReader class to support streaming fake data (pyspark_datasources/fake.py).
• Introduced the _validate_faker_schema function to validate the schema and ensure that the faker library is correctly installed and used (pyspark_datasources/fake.py).

Documentation updates:

• Updated the documentation to include information about the streaming reader and provided an example of how to use it (pyspark_datasources/fake.py). [1] [2]

Tests:

• Added a new test test_fake_datasource_stream to verify the functionality of the streaming reader (tests/test_data_sources.py).

Refactoring:

• Refactored the reader method to use the _validate_faker_schema function and added the streamReader method to the FakeDataSource class (pyspark_datasources/fake.py).