Create more test coverage for ingest pipeline

[weiweishi]

We currently don't have much test coverage for the ingest pipeline, including sources from symphony, SFX, databases, and Kule collections. We'd like to create more tests to validate the ingest is successful, and the metadata are ingested as expected.
As part of this, we will also need to rethink the test dataset #1628. A potential solution is to have separate seed files for UAT where more data is needed for acceptance tests, and one for spec tests.

[weiweishi]

• Have a radioactive record for each of the ingest pipeline. need one for each ingest pipeline.
• Compare test metadata with the data in Solr, check a few fields with Solr queries i.e q=title_t:*
• Compare the number of test records vs the number of item ingested in Solr.