feat(max tokens): increase max tokens from 6k to 15k and add to config

[alissonperez]

What was done?

This pull request introduces the ability for users to configure the maximum tokens for LLM input, increasing the limit from 6000 to 15000. This enhancement allows for greater flexibility in handling larger inputs for language model processing.

How was it done?

• Updated the configuration to include a new parameter INPUT_MAX_TOKENS which defaults to 15000.
• Modified the set_user_config method to allow integer values to be set in the configuration.
• Adjusted the logic in the get_pr_data function to check against the new maximum token limit when processing input.
• Updated documentation to reflect changes in the configuration endpoint.

How was it tested?

• Added unit tests to verify that the INPUT_MAX_TOKENS can be set and retrieved correctly.
• Ensured that the application correctly raises exceptions when the input exceeds the configured maximum tokens.
• Verified that the changes do not break existing functionality by running all existing tests.