Adding handle_long_generation paramters for text-generation pipeline. (#14118)

* Adding `handle_long_generation` paramters for `text-generation` pipeline.

* More error handling

* Fixing tests by dropping tf support on this functionality, it needs

`max_new_tokens` to make it possible to understand user's intent.
Otherwise, `max_length` == `tokenizer.model_max_length` <
input_ids.shape[0].

* Fixing doc ?

* Doc ?

* Remove link from doc.

* Catched an issue on roberta.

* Damn doc.

* Non BC proposal ?

* Cleaning the fix ?

* Finally using only a test override.

* Don't need to modify this.

* Bad print.

tests/test_pipelines_common.py

@@ -143,7 +143,9 @@ class PipelineTestCaseMeta(type):
                     try:
                         tokenizer = get_tiny_tokenizer_from_checkpoint(checkpoint)
                         # XLNet actually defines it as -1.
-                        if (
+                        if model.config.__class__.__name__ == "RobertaConfig":
+                            tokenizer.model_max_length = model.config.max_position_embeddings - 2
+                        elif (
                             hasattr(model.config, "max_position_embeddings")
                             and model.config.max_position_embeddings > 0
                         ):