CampParticipant refactor and related features

[krzys-h]

To be merged after #571, so marking as a draft for now.

Until now, the CampParticipant (previously known as WorkshopUserProfile, see #571) object was basically used as a "camp qualification result" object, created only after the user was qualified. Now, the object will be created immediately after the user registers for the first workshop. This is ensured by the database model - WorkshopParticipant now links to the CampParticipant object. This cleans up some of the funky who-participated-when code, and makes it possible to store per-year user information before they are qualified.

Two features which required this for data storage were implemented (I may split these into separate PRs if needed but I didn't want to make too much of a don't-merge-this-before-X mess, they are separate commits in this PR for now):

• You can now register for the camp even before the program is available. This will create a CampParticipation object, but without any workshops. The message says that the users will be notified when the program appears, but until we have a mailing system this will have to be done manually. New filters were added to the "filtered emails" view to facilitate this - it now differentiates between "users who registered for the camp (clicked the 'im interested' button)" and "users who registered for at least 1 workshop" (the one we had until now). Closes #560
• The cover letter was moved from UserProfile to CampParticipant, to make it a per-year field so that it can be archived. Closes #518
• oh, and I think I accidentally also implemented #206, you can now select multiple CampParticipant objects in Django Admin and mass change the status, similarly to what you could do with Workshops (but the user experience is still not the best since you can't see any of the DataTable stats of people, so we may not want to close that issue just yet)

---

This change is 

[codecov[bot]]

Codecov Report

Merging #572 (24b3a2d) into master (ccacf05) will decrease coverage by 0.50%. The diff coverage is 73.55%.

@@            Coverage Diff             @@
##           master     #572      +/-   ##
==========================================
- Coverage   85.35%   84.84%   -0.51%     
==========================================
  Files          15       15              
  Lines        2362     2402      +40     
==========================================
+ Hits         2016     2038      +22     
- Misses        346      364      +18     

Impacted Files	Coverage Δ
wwwapp/urls.py	100.00% <ø> (ø)
wwwapp/mail_views.py	50.00% <44.44%> (+2.30%)	:arrow_up:
wwwapp/models.py	82.59% <71.87%> (-0.87%)	:arrow_down:
wwwapp/admin.py	82.73% <77.77%> (-1.01%)	:arrow_down:
wwwapp/views.py	80.79% <82.35%> (-0.67%)	:arrow_down:
wwwapp/forms.py	96.07% <100.00%> (ø)
...app/management/commands/populate_with_test_data.py	96.82% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update ccacf05...24b3a2d. Read the comment docs.