Feature/add integration to deepseek

[chakravarthik27]

This pull request includes several significant changes to improve the functionality and robustness of the codebase. The most important changes include adding a new method to extract the final answer from text, handling exceptions more effectively, and updating parameters for model evaluation.

Enhancements to exception handling and text processing:

• langtest/modelhandler/llm_modelhandler.py: Added the extract_final_answer method to process the text and extract the final answer for reasoning models, and modified the predict method to utilize this new method. Additionally, improved exception handling by raising a ValueError with a specific error message.

Updates to model evaluation parameters:

• langtest/utils/custom_types/sample.py: Added a new parameter model_type to the __update_params method to support different model types during evaluation.
• langtest/utils/custom_types/sample.py: Updated the call to load_eval_model.model within the __update_params method to include the new model_type parameter.

[Copilot]

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

Comments suppressed due to low confidence (1)

langtest/utils/custom_types/sample.py:561

• The newly added model_type parameter is passed directly to load_eval_model.model without any validation. Verify that the value from harness_config is compatible with the expected input type of load_eval_model.model.

model, hub, model_type=model_type, **model_parameters

[Copilot]

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.