CI: avoid human error, automatically infer generative models (#33212)

* tmp commit

* move tests to the right class

* remove ALL all_generative_model_classes = ...

* skip tf roberta

* skip InstructBlipForConditionalGenerationDecoderOnlyTest

* videollava

* reduce diff

* reduce diff

* remove  on vlms

* fix a few more

* manual rebase bits

* more manual rebase

* remove all manual generative model class test entries

* fix up to ernie

* a few more removals

* handle remaining cases

* recurrent gemma

* it's better here

* make fixup

* tf idefics is broken

* tf bert + generate is broken

* don't touch tf :()

* don't touch tf :(

* make fixup

* better comments for test skips

* revert tf changes

* remove empty line removal

* one more

* missing one

tests/models/marian/test_modeling_flax_marian.py

@@ -231,7 +231,6 @@ class FlaxMarianModelTester:
 class FlaxMarianModelTest(FlaxModelTesterMixin, unittest.TestCase, FlaxGenerationTesterMixin):
     is_encoder_decoder = True
     all_model_classes = (FlaxMarianModel, FlaxMarianMTModel) if is_flax_available() else ()
-    all_generative_model_classes = (FlaxMarianMTModel,) if is_flax_available() else ()
 
     def setUp(self):
         self.model_tester = FlaxMarianModelTester(self)

tests/models/marian/test_modeling_marian.py

@@ -237,7 +237,6 @@ class MarianModelTester:
 @require_torch
 class MarianModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin, unittest.TestCase):
     all_model_classes = (MarianModel, MarianMTModel) if is_torch_available() else ()
-    all_generative_model_classes = (MarianMTModel,) if is_torch_available() else ()
     pipeline_model_mapping = (
         {
             "feature-extraction": MarianModel,
@@ -871,7 +870,6 @@ class MarianStandaloneDecoderModelTester:
 @require_torch
 class MarianStandaloneDecoderModelTest(ModelTesterMixin, GenerationTesterMixin, unittest.TestCase):
     all_model_classes = (MarianDecoder, MarianForCausalLM) if is_torch_available() else ()
-    all_generative_model_classes = (MarianForCausalLM,) if is_torch_available() else ()
     test_pruning = False
     is_encoder_decoder = False