Reformulation in final_touch

[blegat]

When doing #75, we got hit by jump-dev/MathOptInterface.jl#1980 - the order of calling final_touch between ComplementsVectorizeBridge and VerticalBridge was wrong. So as a workaround, the reformulation were not done in final_touch. This was working because SpecifySetTypeBridge before was doing the formulation in final_touch so we know the bounds already got through. This is the kind of very brittle solution that Claude came up with ^^ Of course, everything would fall apart if a user would create a ComplementsWithSetType constraint directly and the bounds are passed after the bridge are created! So once jump-dev/MathOptInterface.jl#3025 lands, we can do things in final_touch which is what this PR does

Needs

• Fix final_touch order MathOptInterface.jl#3025

[codecov]

Codecov Report

❌ Patch coverage is 95.91837% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 99.69%. Comparing base (28f0ec9) to head (50e536f).

Files with missing lines	Patch %	Lines
src/Bridges/ComplementsVectorizeBridge.jl	95.65%	1 Missing ⚠️
src/Bridges/VerticalBridge.jl	96.15%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##              main      #78      +/-   ##
===========================================
- Coverage   100.00%   99.69%   -0.31%     
===========================================
  Files           12       12              
  Lines          636      656      +20     
===========================================
+ Hits           636      654      +18     
- Misses           0        2       +2     

☔ View full report in Codecov by Harness.
📢 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.