I ran into the same exact problem with several of our consumption-based APIM instances. To make matters even more mysterious, these requests were coming from Azure data center IPs from all around the world.
After a lot of digging, and failed attempts to fix the problem, it turned out this was caused by Availability Tests that are part of Application Insights that we associate with our APIM instances.
If you open the Application Insights resource that is associated to your consumption-based APIM instance in the Azure Portal. Then open the 'Availability' blade under 'Investigate'. You will see where a default availability test, using the /status-0123456789abcdef endpoint. To fix the issue, it looks like you can delete, disable, or update the availability test to hit a valid endpoint. From what I can tell, Application Insights sets this up automatically when you associate it to your APIM instance.
Here is a link to the MS docs for configuring the availability tests:
https://learn.microsoft.com/en-us/azure/azure-monitor/app/monitor-web-app-availability