utilize weakdeps

[aplavin]

wherever possible.
No breaking changes.
And the code itself didn't change at all, it's just github diff isn't the best with file renaming + code moving.

[codecov]

Codecov Report

Attention: Patch coverage is 94.30380% with 9 lines in your changes missing coverage. Please review.

Project coverage is 89.59%. Comparing base (44abdc4) to head (cd92a10).
Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
ext/RecipesBaseExt.jl	94.23%	9 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #154      +/-   ##
==========================================
- Coverage   90.05%   89.59%   -0.46%     
==========================================
  Files          20       20              
  Lines        1156     1153       -3     
==========================================
- Hits         1041     1033       -8     
- Misses        115      120       +5     

Flag	Coverage Δ
unittests	89.59% <94.30%> (-0.46%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[baggepinnen]

Thanks for the PR and sorry for taking a while to get to it.

This should have been a breaking change, since bypmap should have used the function pmap from Distributed, which is no longer loaded by default, while bypmap is still exported by default. However, I notice now that bypmap doesn't actually use pmap and instead uses the ordinary map function. This PR is thus not breaking, but the PR together with fixingthe pmap oversight will break existing usage of bypmap without loading Distributed. This might be fine though, since it has never actually had its intended function before :O

[aplavin]

Hah, interesting :) I didn't even question myself whether something was supposed to use pmap – just removed Distributed and saw that all tests pass.
I guess, much fewer people use Distributed compared to the regular map() or to threads, and even fewer of them want to do pmap over MCM Particles...

[baggepinnen]

and even fewer of them want to do pmap over MCM Particles

It is in the documentation, so it's not unreasonable that people would try to use it.