Fix generate with inputs_embeds as input (#32493)

* I think inputs_embeds has ndim == 3

* fix sequence length catch

* add generate test

* [run-slow]olmo, persimmon, gemma, gemma2, qwen2, llama

* skip whisper

* fix bart test

* more fixes

tests/models/bart/test_modeling_bart.py

@@ -1540,3 +1540,8 @@ class BartStandaloneDecoderModelTest(ModelTesterMixin, GenerationTesterMixin, un
     @unittest.skip
     def test_save_load_fast_init_from_base(self):
         pass
+
+    @unittest.skip(reason="Generate needs input ids")
+    def test_inputs_embeds_matches_input_ids_with_generate(self):
+        # generate only works with input ids for bartforcausalLM
+        pass