ACA Weekly Report initial draft

[jeanconn]

This PR includes code to make a weekly report. This initial draft:

• collects kadi manvrs over the reporting interval
• finds the associated proseco star catalogs (for both catalog data and predictions)
• collects the observed fid, acq, guide data for the observation/manvr
• creates formatted catalog tables with red marking for stars and fids with unexpected values
• checks the observation t_ccd against the predictions
• looks for other kinds of anomalies such as high background or low Kalman star intervals
• makes a formatted table/report with the collected data and jinja2 templates

The in-progress report is available at

https://cxc.cfa.harvard.edu/mta/ASPECT/aca_weekly_report/

To-do items :

• [ ] Put appropriate catalog row and observation text warnings on the detail page
  • [ ] document the catalog row and observation tests (may be self-documenting with above item)
• [ ] fix mica guide stats issues especially missing data and observations where all stars show low fraction
• [ ] create mica fid stats to remove sybase dependency
• [ ] put name hrefs on detail page to link directly (on a guide, fid, acq warning for example)
• [ ] improve titles / templates on detail and main page
• [ ] consider a system for archived weekly reports and navigation between them (for now doing the last week and the last quarter as two separate reports)
• [ ] email warnings on new entries in the table with interesting warnings
• [ ] try to remove duplication in layout/formatting code for the star catalogs and the full top-level table

[jeanconn]

I see my weekly notes say "sprint forward on critical issues" but I don't know which ones were considered critical.

[taldcroft]

Priorities:

• fix mica guide stats issues especially missing data and observations where all stars show low fraction: Not clear what this means, but I get the impression that the tool can show misleading information? That would be bad, but your judgement.
• create mica fid stats to remove sybase dependency: it needs to work on shiny, but maybe this is a non-issue.
• consider a system for archived weekly reports and navigation between them
• email warnings on new entries in the table with interesting warnings

[jeanconn]

I think all of those priorities are fine to address as separate issues. The mica issue is actually a mica problem. The fid_stats should work in a shiny with production perl packages.

[taldcroft]

OK to push those last two off to issues, but they should be addressed with a high priority in order to make this actually useful to us in our normal workflow (of being generally busy and benefitting from push notifications to email).