Evaluate sending quality info to the ELK stack

[gothub]

Evaluate sending quality scores to the ElasticSearch Logstash Kibana (ELK) system used by the Making Data Count project. This would replace the current Solr index used by metadig-engine.

Changes to metadig-engine would involve:

• updating metadig-worker so that data would be sent to a Solr server based on a configuration parameter. This capability would not be removed completely, to allow for flexibility in the future.
• develop an alternative method of parsing the quality reports in order to extract and calculate the quality scores. Currently this is done using a variant of the DataONE d1_index_processor component. A likely alternative would be a Java class using XSLT, then calculating the scores manually.
• metadig-worker would write out a file that contains the quality score info for each pid/authoritative MN/suite id.
• a Filebeat instance would be available on a known port on a k8s cluster machine that would provide info from the data file metadig-worker writes

[gothub]

After reviewing the metadig-engine code, it appears that the d1_index_processor code could still be used for parsing quality reports. A new Java class can be written to used the parsed and calculated info (quality scores, etc) to write out the JSON file that the Filebeat listener will read.