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/plbart/test_modeling_plbart.py

@@ -224,7 +224,6 @@ class PLBartModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMix
     all_model_classes = (
         (PLBartModel, PLBartForConditionalGeneration, PLBartForSequenceClassification) if is_torch_available() else ()
     )
-    all_generative_model_classes = (PLBartForConditionalGeneration,) if is_torch_available() else ()
     pipeline_model_mapping = (
         {
             "feature-extraction": PLBartModel,
@@ -658,7 +657,6 @@ class PLBartStandaloneDecoderModelTester:
 @require_torch
 class PLBartStandaloneDecoderModelTest(ModelTesterMixin, GenerationTesterMixin, unittest.TestCase):
     all_model_classes = (PLBartDecoder, PLBartForCausalLM) if is_torch_available() else ()
-    all_generative_model_classes = (PLBartForCausalLM,) if is_torch_available() else ()
     test_pruning = False
     is_encoder_decoder = False