develop methods to access data via a store

[jeanetteclark]

to run data quality checks, python checks are going to need to be able to access data files directly on disk. This should be written in such a way that check code would undergo no changes were the method to access the files to change.

my plan is to implement this as an interface with the following components:

• ObjectStore interface that defines get_object which will return a stream
• HashStore implementation of the ObjectStore interface which accesses a hashstore backend
• StoreManager that manages the backend store

so the calling code in the checks would look something like this, where store_configuration is a hashmap passed over from Java with configuration information including the store type

store_manager = StoreManager(store_configuration)
obj = store_manager.get_object("pid")

[jeanetteclark]

this work is mostly done but still needs implementation in check code. Leaving it on branch feature-store-interface and issue open until the corresponding metadig-checks (re-writing data checks to use it) and metadig-engine (add methods to pass pids to checks) is completed

[jeanetteclark]

implementation in check code has been completed, see metadig-checks branch feature-data-quality to see the drafted data.suite and it's corresponding checks. this is all just awaiting full integration testing on k8s and review