Enable ESS calculation in metric calc

[acordonez]

Summary of changes and related issue

• Add check_ess option
• Update ESS code in metric_calc to run on data with extra dims like sim
• Update unit tests as needed
• • metric_calc_param_validator.py: 83% coverage
• • processor_utils.py: 81% coverage
• • metric_calc.py: 90% coverage (across multiple test files)
• A variable name bugfix from the confidence interval changes

Link to corresponding Jira ticket(s)

AE-1412

Testing

• Unit tests written for new/modified code (goal: 80% coverage)
  • All public functions have unit tests
  • Functions that must produce specific values have unit tests
• Verified that notebooks utilizing affected functions still work
• Appropriate manual testing completed
• Advanced Testing label added to PR if this PR makes major changes to the codebase

How to Test

Here's a notebook set up with several tests to check that the ESS test runs, or doesn't, as expected:
ess_test.ipynb

Also verify that unit tests are working in CI.

Documentation

• Complex code includes comments explaining logic
• Function documentation created (docstrings)

Code Quality

• Follows PEP8 naming and style conventions
• Helper functions prefixed with underscore _
• Linting completed and all issues resolved
• Does not replicate existing functionality
• Aligns with general coding standards of existing codebase
• Code generalized for multiple uses (unless too complex/time-intensive)

Review Process

• PR review instructions provided:
  • Type of review requested (scientific/technical/debugging)
  • Level of review detail needed

Administrative Reminders

• Jira ticket moved to "Review" when PR created
• Jira ticket will be moved to "Done" when complete
• PR branch will be deleted after merge

---

[acordonez]

@claalmve and @neilSchroeder looking for both technical and methods review - I want to make sure I haven't changed the underlying methodology for the effective sample size, just extended it to run over datasets with more dimensions.

[neilSchroeder]

A few small things

[claalmve]

This PR looks great! Tested the notebook, everything looks well. I only left one comment about my confusion on the docstring, otherwise no notes!

[claalmve]

Beautiful!

[claalmve]

How come it's 1 - significance level and not just significance level?

[acordonez]

We could use confidence level as a variable rather than alpha, I just thought of alpha first. Alpha's the variable for the probability of being in the tails of the distribution, outside of the confidence interval.

[acordonez]

Updated the alpha variable description since I've been getting different feedback on the significance level part - leaving that out.