Avoid collisions in writing metrics via 2 APIs - azureml + mlflow (#20837)

* Avoid collisions in writing metrics via 2 APIs - azureml + mlflow

MLflow tracking API is enabled by default in AzureML and HF MLflow integration is more fully featured. I'd remove the AzureML integration but leaving the current behavior for backwards compatibility (though it should really be removed)

* Trigger CI

src/transformers/integrations.py

@@ -518,7 +518,7 @@ def run_hp_search_wandb(trainer, n_trials: int, direction: str, **kwargs) -> Bes
 
 def get_available_reporting_integrations():
     integrations = []
-    if is_azureml_available():
+    if is_azureml_available() and not is_mlflow_available():
         integrations.append("azure_ml")
     if is_comet_available():
         integrations.append("comet_ml")