feat: npsim 1.5.0, patch dd4hep warning for Geant4TVUserParticleHandler

[veprbl]

eic/eic-spack#896

[Copilot]

Pull request overview

Updates the pinned eic/eic-spack revision used by this repository, in order to pick up upstream changes (per linked eic-spack PR #896).

Changes:

• Bump EICSPACK_VERSION commit hash in eic-spack.sh.

[github-actions]

Capybara summary for PR 257

• rec_dis_10x100_minQ2=1000_epic_craterlake
• rec_dis_10x100_minQ2=1_epic_craterlake
• rec_dis_18x275_minQ2=1000_epic_craterlake
• rec_dis_18x275_minQ2=1_epic_craterlake
• rec_dis_5x41_minQ2=1_epic_craterlake
• rec_e_1GeV_20GeV_epic_craterlake
• rec_pi_1GeV_20GeV_epic_craterlake
• sim_dis_10x100_minQ2=1000_epic_craterlake
• sim_dis_10x100_minQ2=1_epic_craterlake
• sim_dis_18x275_minQ2=1000_epic_craterlake
• sim_dis_18x275_minQ2=1_epic_craterlake
• sim_dis_5x41_minQ2=1_epic_craterlake
• sim_e_1GeV_20GeV_epic_craterlake
• sim_pi_1GeV_20GeV_epic_craterlake
  ^{_{Last updated 2026-04-27T23:23-04:00 a65a4e7}}

[veprbl]

Continuing the last discussion in eic/epic#1081

# capybara capy rev head --owner eic --repo containers --artifact-name rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root
# capybara capy rev 3efd159b102646ac932a4f365fc7e27e6839af01 --owner eic --repo containers --artifact-name rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root
# du -hs 2026-04-26T12-49-43+00-00_3efd159b102646ac932a4f365fc7e27e6839af01/rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root 2026-04-28T00-48-39+00-00_a65a4e733a72f81d3ab884da981648740cdeb26a/rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root
154M	2026-04-26T12-49-43+00-00_3efd159b102646ac932a4f365fc7e27e6839af01/rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root
276M	2026-04-28T00-48-39+00-00_a65a4e733a72f81d3ab884da981648740cdeb26a/rec_dis_18x275_minQ2=1000_epic_craterlake.edm4eic.root

Looks like an almost 2x file size increase. I'm not sure we can afford this. I wonder if we can define a more optimal tracking volume?

[DraTeots]

So it is basically the size of MCParticles that describe all shose tiny shower details.
I didn't see this on our files, but ours are very simple e+K+lambda compared to something like NC/CC DIS.

There is another dimension to optimize. MCParticle is saved if it was born in tracking volume and its energy is 1MeV or more. Can we also tweak this 1MeV? And especially for different part of volumes?

[wdconinc]

What is the path towards progress here? Should we yank npsim 1.5? Should we yank the epic TV extension? We are now in the unfortunate condition that the head of the default branches in those repositories are conspiring to be factor 2 inefficient. And the only place where that's documented is here in a third repository.

[veprbl]

npsim 1.5.x is not bad or broken, just not suitable for large scale productions. We meet 3PM ET to discuss possible ways to optimize this.