fix: add batching support for BanCompetitors to handle long input text

[abdokaseb]

Change Description

Previously, the BanCompetitors class failed to work with long text because it attempted to process the entire input text in a single request. This caused incomplete processing when the input text exceeded the model's token limit as it truncates input text.

This PR introduces batching logic to split the input text into manageable chunks before sending it to the model. This ensures that BanCompetitors works correctly even with long inputs and models that have lower token limits. Also added a test case for long text that fail for old code and now it pass correctly.

Assumptions:
For fast processing I used approximate token count equation that I already mentioned in the code 1 word ~ 4 characters ~ 2 tokens

Issue reference

N/A – discovered during usage with large text inputs. Please let me know if you'd like me to open a tracking issue.

Checklist

• I have reviewed the contribution guidelines
• My code includes unit tests
• All unit tests and lint checks pass locally
• My PR contains documentation updates / additions if required