checkpoint model on first step callback (#2906)

* checkpoint model on first step callback

* remove debug

* add test cases; update existing tests not to save on first step

* move test out of solo

* delete

* default to False

* typo

tests/e2e/test_qwen.py

@@ -59,6 +59,7 @@ class TestE2eQwen:
                 "bf16": "auto",
                 "tf32": True,
                 "gradient_checkpointing": True,
+                "save_first_step": False,
             }
         )