Why does an @id node identifier referencing the top node stop Google’s Rich Results test detecting a logo?

For example it says "Page not eligible for rich results known by this test" for the following, but if you remove the "about" property then it successfully detects the logo.

Is this a bug? Is there a way to work around it besides removing all @id references to the top node? In our real JSON-LD I also use "author": {"@id": "https://acme.com/"} for various offered SoftwareApplications, which I’d rather not remove.

{     "@context": "http://schema.org/",     "@id": "https://acme.com/",     "@type": "Corporation",     "url": "https://acme.com/",     "mainEntityOfPage": "https://acme.com/",     "name": "Acme",     "image": "https://acme.com/logo.png",     "logo": "https://acme.com/logo.png",     "brand": {         "@id": "https://acme.com/#brand",         "@type": "Brand",         "logo": "https://acme.com/logo.png",         "name": "Acme",         "url": "https://acme.com/"     },     "subjectOf": {         "@id": "https://acme.com/about",         "@type": "AboutPage",         "mainEntityOfPage": "https://acme.com/about",         "name": "About Acme",         "url": "https://acme.com/about",         "about": {             "@id": "https://acme.com/"         }     } }