Add: mlflow for experiment tracking

src/axolotl/core/trainer_builder.py

@@ -738,7 +738,9 @@ def build(self, total_num_steps):
             False if self.cfg.ddp else None
         )
         training_arguments_kwargs["group_by_length"] = self.cfg.group_by_length
-        training_arguments_kwargs["report_to"] = "wandb" if self.cfg.use_wandb else None
+        training_arguments_kwargs["report_to"] =  (
+            "wandb" if self.cfg.use_wandb else 'mlflow' if self.cfg.use_mlflow else None 
+        )
         training_arguments_kwargs["run_name"] = (
             self.cfg.wandb_name if self.cfg.use_wandb else None
         )