Add option to store the Hessian in sparse format for R2N

[MaxenceGollier]

#194

@MohamedLaghdafHABIBOULLAH @dpo let me know what you think when you have some time, thank you !

I have tested this locally with CUTEst, I can not test it with bpdn since it is a FirstOrderModel and can therefore not give me the Hessian in sparse format... Should I add CUTEst as a dependency and do my unit tests with it or are there other models in RegularizedProblems that are not FirstOrderModels

[codecov]

Codecov Report

Attention: Patch coverage is 39.39394% with 20 lines in your changes missing coverage. Please review.

Project coverage is 58.82%. Comparing base (e0f214d) to head (5ccacbf).
Report is 88 commits behind head on master.

Files with missing lines	Patch %	Lines
src/R2NModel.jl	13.33%	13 Missing ⚠️
src/R2N.jl	61.11%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #195      +/-   ##
==========================================
- Coverage   61.53%   58.82%   -2.71%     
==========================================
  Files          11       15       +4     
  Lines        1292     1909     +617     
==========================================
+ Hits          795     1123     +328     
- Misses        497      786     +289     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

Here are the
demos-results

[github-actions]

Here are the
demos-results

[dpo]

Sorry for the delay. I remember that we talked about using QuadraticModels instead of R2NModel. What were the blocking factors?

[MaxenceGollier]

Sorry for the delay. I remember that we talked about using QuadraticModels instead of R2NModel. What were the blocking factors?

Type instabilities if i remember well.