fastavro is no longer used by default

[mparry]

Description

There seems to have been a bug introduced in #470 that meant that, unless reader schemas are manually specified in the AvroConsumer constructor, attempts to use fastavro will always fail. (That is, the reader_schema_obj variable will be None, the attempt to call to_json() on it will throw, and the catch-all error handling will mean that it silently falls back to plain Python avro.)

As well as being rather a lot slower, this alters deserialisation behaviour in some cases.

This seems to have first been released in 1.0.0.

Checklist

Please provide the following information:

• confluent-kafka-python and librdkafka version (confluent_kafka.version() and confluent_kafka.libversion()):
• Apache Kafka broker version:
• Client configuration: {...}
• Operating system:
• Provide client logs (with 'debug': '..' as necessary)
• Provide broker log excerpts
• Critical issue