[AIST-QA] Fix memory leak issue in rdf_loader

[max-krichenbauer]

Description

In the constructor of RDFLoader class, when umodel can not initialized, an error is shown and the constructor returns after showing an error message. In this case however, the umodel variable is not deleted which causes a memory leak.
This fix adds delete umodel; to the scope when constructor terminate prematurely.

This contribution is made by AIST ( https://www.aist.go.jp ) based on static code analysis with klocwork (Perforce Software).

Checklist

• Required by CI: Code is auto formatted using clang-format
• Extend the tutorials / documentation reference
• Document API changes relevant to the user in the MIGRATION.md notes
• Create tests, which fail without this PR reference
• Include a screenshot if changing a GUI
• While waiting for someone to review your request, please help review another open pull request to support the maintainers

[rhaschke]

Thanks for pointing out this issue. I proposed a cleaner solution for this, avoiding new/delete completely in #1830. Closing here.

[max-krichenbauer]

Thank you for the review and for proposing a better way to fix it!
Generally though, it would be much appreciated if you would allow us to modify this PR to meet your expectations instead of closing it and opening a new PR, because these PRs are used internally by our project management to evaluate if our efforts are effective and valued by the community.

[rhaschke]

My policy for such minor fixups is to apply them directly (particularly if I opened the source editor anyway to see the context). Usually I push those changes to the PR branch to keep them with the original PR. However, your PR was filed on an organizational account and github forbids pushing to those.
If you prefer adapting the PR yourself, I can try to remember this in future.

[max-krichenbauer]

Thank you! I understand that it's probably annoying for you to go back-and-forth with PRs instead of just fixing things immediately, but for us it's hard to justify our work if our PRs are apparently being rejected.
(Of course, if the content of the contribution is not considered useful, feel free to reject PRs).

We're trying our best to be as responsive as possible, so if it's not too much of a hassle, could we be allowed to modify this PR?

[rhaschke]

Feel free to just fetch my additional changes in #1830 on top of your PR here.

[max-krichenbauer]

Thanks a lot!
Just one minute...

[rhaschke]

Sorry, I used the wrong type names. I hope, #1830 is fixed now (wait for Travis to finish).

[max-krichenbauer]

Thanks! The checks on #1830 passed, so I applied the changes to this PR.

Thank you again so much for your understanding!

[rhaschke]

Thanks. Please remove the unrelated cleanup in end_effectors_widget.cpp.

[max-krichenbauer]

Got it. Sorry for the misunderstanding.

[rhaschke]

Fine with me if Travis is happy as well.