Differentiate conic model through bridges

[blegat]

See #167

Tests are failing because of #170

[codecov]

Codecov Report

Merging #184 (4647af6) into master (ab872dc) will increase coverage by 2.55%.
The diff coverage is 98.34%.

@@            Coverage Diff             @@
##           master     #184      +/-   ##
==========================================
+ Coverage   86.58%   89.13%   +2.55%     
==========================================
  Files           7        8       +1     
  Lines         872      911      +39     
==========================================
+ Hits          755      812      +57     
+ Misses        117       99      -18     

Impacted Files	Coverage Δ
src/conic_diff.jl	97.98% <97.87%> (+1.31%)	⬆️
src/bridges.jl	100.00% <100.00%> (ø)
src/diff_opt.jl	96.75% <100.00%> (+5.33%)	⬆️
src/moi_wrapper.jl	81.15% <100.00%> (+0.75%)	⬆️
src/product_of_sets.jl	100.00% <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 ab872dc...4647af6. Read the comment docs.

[JinraeKim]

See #167

Tests are failing because of #170

I wonder if this bridge automatically transforms convex optimisation problems into corresponding conic problems to perform the differentiation through the optimisation problem. Is it right?
+) is this bridge automatically gonna warn if a given problem is non-convex? Or, isn't it gonna work?

[blegat]

DCP will be a bridge once jump-dev/MathOptInterface.jl#846 is done

[JinraeKim]

DCP will be a bridge once jump-dev/MathOptInterface.jl#846 is done

I'll take a look at it. Thank you!