dangerouslyForceOverride

This is a combination of dangerouslyReset and override that reduces the likeliness of a race condition between the two calls.

This is dangerous because it can introduce consistency issues that will be much harder to debug. For example, it could hide the fact that feature flags are read before you set the values you want to use everywhere. It could also cause a workflow to suddenly have different feature flags for behaviors that were configured with different values before.

It returns a string that contains the feature flags that were accessed before this call (or between the last call to dangerouslyReset and this call). If you are using this method, you do not want the hard crash that you would get from using dangerouslyReset and override separately, but you should still log this somehow.

Please see the documentation of dangerouslyReset for additional details.