Changes to presolve initial sweep

[fwesselm]

• Remove std::vector<bool>
• Use HighsCDouble for activity bounds (similar to HighsLinearSumBounds)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.77%. Comparing base (57e5065) to head (b196d7a).
⚠️ Report is 92 commits behind head on latest.

Additional details and impacted files

@@           Coverage Diff           @@
##           latest    #3210   +/-   ##
=======================================
  Coverage   72.77%   72.77%           
=======================================
  Files         444      444           
  Lines      107464   107469    +5     
  Branches    17288    17290    +2     
=======================================
+ Hits        78208    78213    +5     
  Misses      28979    28979           
  Partials      277      277           

☔ 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.

[jajhall]

Thanks. I thought it was OK to compute infinite implied bounds explicitly, rather than tracking them with the HighsBool vectors implied_row_*_finite, but fair enough

[fwesselm]

Thanks. I thought it was OK to compute infinite implied bounds explicitly, rather than tracking them with the HighsBool vectors implied_row_*_finite, but fair enough

I think using HighsCDouble is beneficial if the activity bounds are large.

kHighsInf cannot be converted to HighsCDouble, at least I ran into issues last time I tried, hence the additional indicator vectors.

[jajhall]

kHighsInf cannot be converted to HighsCDouble, at least I ran into issues last time I tried, hence the additional indicator vectors.

Very true

[fwesselm]

kHighsInf cannot be converted to HighsCDouble, at least I ran into issues last time I tried, hence the additional indicator vectors.

Very true

However, I can still try to get rid of the HighsBool vectors. I will update the PR.