Majority Vote and Weighted majority vote does not group by equivalent answers?

Hi!

Thanks for open sourcing the implementation of DeepThink. I was looking into the implementation of `simple_majority_vote` and `weighted_majority_vote` [here](https://github.com/anirudhb11/deepconf/blob/main/deepconf/utils.py#L67-L90), and one thing I noticed is that the implementation does not group equivalent answers together. So for a given question, if the model answers `0.5` or `\frac{1}{2}`, these will be counted as two separate answers, when in reality they are the same answer.

Is this intended?

Thanks!

Start in deepconf/utils.py at lines 67-90 and inspect simple_majority_vote and weighted_majority_vote. Reproduce the reported case with answers such as 0.5 and \frac{1}{2}; done means equivalent answer representations are grouped consistently by both voting functions, with the behavior covered by tests.

Written by the indexing model from the issue text.