Upgrading DiffOpt to MOI 0.10

[matbesancon]

Still a lot of errors I haven't figured out yet

• Replace MatrixOptInterface by MOI.Utilities.MatrixOfConstraints

Closes #150

[matbesancon]

starts to tackle #150

[codecov]

Codecov Report

Merging #150 (d9cc2e6) into master (d60c53f) will increase coverage by 0.23%.
The diff coverage is 92.30%.

@@            Coverage Diff             @@
##           master     #150      +/-   ##
==========================================
+ Coverage   85.84%   86.08%   +0.23%     
==========================================
  Files           6        7       +1     
  Lines         756      769      +13     
==========================================
+ Hits          649      662      +13     
  Misses        107      107              

Impacted Files	Coverage Δ
src/moi_wrapper.jl	78.83% <64.70%> (+2.02%)	⬆️
src/conic_diff.jl	96.66% <93.33%> (-3.34%)	⬇️
src/diff_opt.jl	87.29% <100.00%> (-0.52%)	⬇️
src/jump_moi_overloads.jl	62.96% <100.00%> (-0.46%)	⬇️
src/product_of_sets.jl	100.00% <100.00%> (ø)
src/quadratic_diff.jl	99.12% <100.00%> (-0.01%)	⬇️
src/utils.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 d60c53f...d9cc2e6. Read the comment docs.

[blegat]

There is an issue on Julia v1.0:

• LazyArrays is v0.16 or earlier for Julia v1.0, these version don't support StaticArrays v1.0
• MathOptSetDistances (MOSD) requires StaticArrays v1.0

We have 2 options: either relax the requirement in MOSD or drop support for old Julia versions.

[matbesancon]

we should just ditch Julia 1.0 for DiffOpt, it's not directly a solver and not registered yet, 1.6 can become the new lower bound

[matbesancon]

I set julia 1.6 as the lower bound for now on this

[blegat]

Looks good to merge, let's track the issue of the dirty flag in #152 and not keep this PR open for this.