fix: detect multivariate differences between environments

[10done]

Thanks for submitting a PR! Please check the boxes below:

• I have read the Contributing Guide.
• I have added information to docs/ if required so people know about the feature.
• I have filled in the "Changes" section below.
• I have filled in the "How did you test this code" section below.

Changes

Fixes #7619

• Adds hasMultivariateChange() to compare multivariate_feature_option + percentage_allocation per environment
• Uses it in CompareEnvironments for changed/unchanged classification and row fade behavior
• Uses it in CompareFeatures so MV-only differences are not shown as identical
• Adds unit tests in compare-multivariate-utils.test.ts

How did you test this code?

I have tested the code manually and added screenshots below and also added unit tests for the same.

Before

No flag differences / rollout_test in Unchanged (Dev 0% vs Prod 100%).

After

rollout_test under Changed Flags.

[vercel]

@10done is attempting to deploy a commit to the Flagsmith Team on Vercel.

A member of the Team first needs to authorize it.

[gemini-code-assist]

Code Review

This pull request introduces a utility function hasMultivariateChange to detect differences in multivariate feature allocations when comparing environments and features, along with corresponding unit tests. The feedback suggests defensively converting multivariate_feature_option to a number during normalization to prevent potential NaN values from causing unstable sorting.