Adding skip_special_tokens=True to FillMaskPipeline (#9783)

* We most likely don't want special tokens in this output.

* Adding `skip_special_tokens=True` to FillMaskPipeline

- It's backward incompatible.
- It makes for sense for pipelines to remove references to
special_tokens (all of the other pipelines do that).
- Keeping special tokens makes it hard for users to actually remove them
  because all models have different tokens (<s>, <cls>, [CLS], ....)

* Fixing `token_str` in the same vein, and actually fix the tests too !

src/transformers/pipelines/fill_mask.py

@@ -179,10 +179,10 @@ class FillMaskPipeline(Pipeline):
                 tokens = tokens[np.where(tokens != self.tokenizer.pad_token_id)]
                 result.append(
                     {
-                        "sequence": self.tokenizer.decode(tokens),
+                        "sequence": self.tokenizer.decode(tokens, skip_special_tokens=True),
                         "score": v,
                         "token": p,
-                        "token_str": self.tokenizer.convert_ids_to_tokens(p),
+                        "token_str": self.tokenizer.decode(p),
                     }
                 )