Fix/fake sample count het

[Jiya873]

This PR addresses part of the existing issues #775 with sample-level heterozygosity handling by:

• Refactoring the heterozygosity functionality in the AnophelesDataResource class to expose a public API instead of relying solely on the private _sample_count_het method.
• Unit tests (with monkey-patching) for the heterozygosity method will be added to ensure the method returns the expected results.
• Setting up a dummy logging mechanism to avoid dependency issues during testing.

The handling of ROH and cohort-level comparisons remains suboptimal and will be addressed in a separate effort. This change focuses solely on improving heterozygosity accessibility for single samples.

[jonbrenas]

Thank you @Jiya873. This is great but I think there are several steps that need to be done before we get to the public function:

1. All of the ROH and het functions currently in anopheles.py need to be moved to their own file under anoph/
2. The tests in tests/anoph/ need to be structured similarly to what exists for other functions
3. The public functions need to be updated to meet with the newer requirements, including surveillance_use_only and unrestricted_access that are currently being implemented (see Add unrestricted_use_only and surveillance_use_only constructor params #724), and we need to make sure that the functions make sense for the analyses that they are connected too. In this case, we need a function that returns the heterozygosity for a single sample, another one that works at the cohort(s) level, and at least one plotting function.

That's a lot of work (a lot of which is dependent on other things to complete first) so there is no pressure to proceed with this PR quickly.

[leehart]

Looks like this PR is still in progress.

[jonbrenas]

I think it has been abandoned.

[leehart]

Thanks @jonbrenas . If there's no response after a short time (say, September) then I guess we'll close this. This PR can always be resurrected, if needs be.