[ATLASDPD-2027] fix memleak in VertexIterativeFitMergingTool

There is a memory leak in the iterative vertex merging step (fitSeedVertexCluster) of the VertexIterativeFitMergingTool. If more than one merging step happens, the intermediate merged vertices would leak.

This MR should fix most of this. Local testing on a data file shows the estimated leak rate drop from 350kb/evt to 35kb/evt, and the vmem and RSS now stop growing after the first ~100 events.

CC @mvanadia @kbenkend @pgadow

This merge request affects 1 package:

• PhysicsAnalysis/AnalysisCommon/LeptonTaggers

This merge request affects 2 files:

• PhysicsAnalysis/AnalysisCommon/LeptonTaggers/LeptonTaggers/VertexIterativeFitMergingTool.h
• PhysicsAnalysis/AnalysisCommon/LeptonTaggers/src/VertexIterativeFitMergingTool.cxx

Adding @yoyamagu ,@fcardill as watchers

added Analysis main review-pending-level-1 labels

pss vs walltime in local testing. "new" is this MR, "old" is muon5 in main, "muon1" is a cross-check using a muon1 job. The slope reported in ATLASDPD-2027 seems to be removed. The longer walltime in "new" is not a slowdown, I just ran more events in that test to ensure we don't see a significant slope down the line.

CI Result SUCCESS (hash d8425150)

	Athena
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 5943] (remote access info)

I only have one minor comment. But as this is memory related i would like L2 to have a look as well.

Cheers L1

added review-pending-level-2 label and removed review-pending-level-1 label

resolved all threads

added review-approved label and removed review-pending-level-2 label

Tim - L2

merged

mentioned in commit 0b63a2d2