Check the cause of test failures with OSQP / MOMA

[exaexa]

The problem here:
https://gitlab.lcsb.uni.lu/lcsb-biocore/COBREXA.jl/-/jobs/309330

MOMA: Test Failed at /builds/lcsb-biocore/COBREXA.jl/test/analysis/minimize_metabolic_adjustment.jl:13
  Expression: isapprox(moma["BIOMASS_Ecoli_core_w_GAM"], 0.06214149238730545, atol = TEST_TOLERANCE)
   Evaluated: isapprox(0.05885376722059405, 0.06214149238730545; atol = 9.999999999999999e-6)
Stacktrace:
 [1] macro expansion
   @ /builds/lcsb-biocore/COBREXA.jl/test/analysis/minimize_metabolic_adjustment.jl:13 [inlined]
 [2] macro expansion
   @ /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.6/Test/src/Test.jl:1151 [inlined]
 [3] top-level scope
   @ /builds/lcsb-biocore/COBREXA.jl/test/analysis/minimize_metabolic_adjustment.jl:2
[ Info: analysis/minimize_metabolic_adjustment.jl done in 12.43s

...was kinda quickfixed in PR #484 by simply increasing the test tolerance. A better explanation of what actually happened is needed.

Likely the freedom in the model just allows the results to come from a relatively big range, at that point we wouldn't need to care a lot but it would be better checked by at least 1 other person. :D

[exaexa]

Closed by #486