Johnfreeman/issue517 autofill dunedaq db path

[jcfreeman2]

This PR is very much in the spirit of #352. In a nutshell, rather than Software Coordination needing to manually add a repo's addition to DUNEDAQ_DB_PATH to the repo's package.py file, make-release-repo.py has been updated to check for OKS data and schema files in a repo and, should they exist, to update package.py accordingly by substituting the XDBPATHX token. Note that DDPFD_DEV_260204_A9 is built using this feature branch of daq-release and that it recreates the performance of the extended integration tests (https://github.com/DUNE-DAQ/daq-release/actions/runs/21680817046) and the classic integration tests (https://github.com/DUNE-DAQ/daq-release/actions/runs/21686107769).

[andrewmogan]

I think this all looks good, but I want to check something before approving. If I create a local build area based off DDPFD_DEV_260204_A9 and compare the list of packages in DUNEDAQ_DB_PATH to the same list in an area based on the latest nightly (NFD_DEV_260205_A9), I see slight differences. Namely, restcmd, dfmodules, and iomanager are only in DDPFD_DEV_260204_A9's DUNEDAQ_DB_PATH, while crtmodules and asiolibs are only in NFD_DEV_260205_A9's. Is this expected?

[jcfreeman2]

Good cross check. To explain:

• crtmodules and asiolibs in hindsight didn't in fact need to be added to DUNEDAQ_DB_PATH, though of course I manually added them in JCF: Issue #518: add snbmodules, asiolibs and crtmodules to DUNEDAQ_D… #519. They started working not because I added them to DUNEDAQ_DB_PATH but because Kurt fixed the way the accessed daqsystemtest's OKS XML files, Replaced the relative path to the daqsystemtest/example-configs.data.xml file... crtmodules#29 and Replaced the relative path to the daqsystemtest/example-configs.data.xml file.… asiolibs#35
• As far as the new additions: restcmd's OKS XML file certainly looks like it could be used at some point in the future, though perhaps it's cruft? And iomanager and dfmodules use their OKS XML files just for unit testing, though I'm not sure there's an algorithmic way to determine "authorial intent" behind the reason OKS XML files are in a repo.

[andrewmogan]

Thanks, that helps.

With that, I approve.