Concept data does not get exported in incremental mode in certain situations

[gowthamrao]

In current workflow, concept set details (especially concept table) is download as the last step. This creates scenarios when this data is not extracted.

To reproduce:

1. Run Cohort Diagnostics with default parameters in incremental mode
2. When concept set diagnostics is complete (for example when incidence rate is computed) interrupt and terminate the job
3. Restart in incremental mode

No concept table is extracted. This is because resolved concepts is temporary table

[gowthamrao]

This is a less optimal design especially when CohortDiagnostics is running large jobs such as phenotype library. Interruptions happen, then on resumption - no concept table is extracted.

The reason for the separation in the past is to extract the concepts that are part of visit_context diagnostics.

Potential solution: There are certain concepts that we may consider to be always needed. e.g. concepts in visit domain, gender domain. We can either extract them all as default for every project, or keep a copy of these extremely small csv files in inst folder of package. That way, we do not need to extract concepts during visit context, and all concept ids that need to be extracted are extracted during concept set diagnostics.