Fixing NER pipeline for list inputs.

[Narsil]

What does this PR do?

• Changes TokenArgumentHandler(*args) into (inputs) signature to follow __call__ signature.
• Fixes the bug.
• Backward compatible for single sentences
• Not backward compatible for multiple sentences, but it "worked" only for same length sentences in tokens (the result was bogus as it contained only the first sentence)
• This make NER not pass any batching to the model, which is not in line with other pipelines, however this is what was done beforehand. And not all pipelines support batching (and even batching is counterproductive in a lot of cases because the user cannot control number of tokens from raw strings).
• All slow tests now pass, argparser test was updated.

Fixes #10168

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

@LysandreJik

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors which may be interested in your PR.

[LysandreJik]

LGTM, thank you for taking care of it @Narsil!