I am aware of the documented alternative recipe phrasing that you've demonstrated. The Singularity recipe that I provided in the bug report is much smaller than the complete recipe that I actually use. The complete receipt cannot be optimized in the manner suggested.
I had already indicated that the problem doesn't occur on Debian GNU/Linux installed on bare metal in my original report: there was no need to check that, but thanks. I do not understand why the bug is being closed as "Done". Even if you choose not to investigate the problem yourself (and I don't think anyone is forcing you to do so?), doesn't it seem it unethical to you to close the bug as effectively fixed when it clearly hasn't been? Doing so deprives others of the opportunity to suggest investigative steps that I could try in order to root-cause the problem.
