Option to use raw SQL to improve performance of site monthly metrics with large #s of StudentModules

Change description

Tries to resolve performance issues running backfill_figures_monthly_metrics or even just running the site monthly metrics each month by schedule, in cases where there is a large number of students and StudentModules. I was not able to complete a month on a site with 95k students and ~7 million StudentModule records without pegging system memory and making the whole site unresponsive.

Using a raw SQL query provides performance benefits over using Python/Django... The main bottleneck with Django is the calculation of distinct() student_ids on the QuerySets. Raw SQL DISTINCT() function is significantly more performant.

This PR provide a use_raw parameter to pipelines.site_monthly_metrics.fill_month and "upstream" functions and management commands.

Includes tests, though the actual SQL statement has to be changed to accommodate the differences in sqllite.

Type of change

[x] Bug fix (fixes an issue)
[ ] New feature (adds functionality)

Checklists

Development

[x] Lint rules pass locally
[x] Application changes have been tested thoroughly
[x] Automated tests covering modified code pass

Security

[x] Security impact of change has been considered
[x] Code follows company security practices and guidelines

Code review

[x] Pull request has a descriptive title and context useful to a reviewer. Screenshots or screencasts are attached as necessary
[x] "Ready for review" label attached and reviewers assigned
[ ] Changes have been reviewed by at least one other contributor
[ ] Pull request linked to task tracker where applicable

appsembler / figures