What this PR does / why we need it:
We've recently seen a few edge cases where some referenced resources were already successfully deleted but the deletion flow was not finished completely. Upon retry the deletion was now stuck since the supposedly still existing resources could not be found anymore.
This PR makes the deletion more resilient against these cases.
Special notes for your reviewer:
Release note:
Deletion will now continue if non-gardener-managed loadbalancers and subnets were already deleted beforehand
How to categorize this PR?
/area robustness /kind enhancement /platform azure
What this PR does / why we need it: We've recently seen a few edge cases where some referenced resources were already successfully deleted but the deletion flow was not finished completely. Upon retry the deletion was now stuck since the supposedly still existing resources could not be found anymore. This PR makes the deletion more resilient against these cases.
Special notes for your reviewer:
Release note: