[hist] extend LIST option to support CONT1 CONT2 CONT3

[ferdymercury]

This Pull request:

Changes or fixes:

fyi @bgrube

Fixes #20858

Follow-up of d4b273b

Checklist:

• tested changes locally
• updated the docs (if necessary)

[couet]

The LIST option retrieves the contours generated by the CONT0 option. I am not sure the other algorithms work the same way as CONTO and therefore the contours might not be generated as CONTO does it.

[ferdymercury]

I verified that it works with this test script:

   auto c1 = new TCanvas("c1","c1",600,400);
   auto hcont1 = new TH2F("hcont1","Option CONT1Z LIST example ",40,-4,4,40,-20,20);
   float px, py;
   for (Int_t i = 0; i < 25000; i++) {
      gRandom->Rannor(px,py);
      hcont1->Fill(px-1,5*py);
      hcont1->Fill(2+0.5*px,2*py-10.,0.1);
   }
   hcont1->Draw("CONT1Z LIST");
   gPad->Update();
   new TCanvas();
    ((TGraph*)((TList*)((TObjArray*)gROOT->GetListOfSpecials()->FindObject("contours"))->At(0))->First())->Draw("AL");

[couet]

Ok LGTM

[github-actions]

Test Results

    23 files      23 suites   3d 21h 35m 9s ⏱️
 3 813 tests  3 813 ✅ 0 💤 0 ❌
80 376 runs  80 376 ✅ 0 💤 0 ❌

Results for commit 4dc9352.

♻️ This comment has been updated with latest results.

[ferdymercury]

gentle ping (CI is green, good to merge from my side)

[dpiparo]

thanks. I am preparing the backports.

[hageboeck]

gentle ping (CI is green, good to merge from my side)

Sorry, it took super long because there were lots of CI instabilities in the past days. Had nothing to do with the PR, but wanted to be super sure.