Fix bugs with large values in multimodel regression tests

[stefsmeets]

Description

This PR changes the regridder to a Nearest scheme to avoid interpolation errors on masked arrays. For more details see #1014 . I also added a check for large values, which revealed another dataset with a large unmasked value: ESMValGroup/ESMValTool_sample_data#24 that I would like to include in this PR as well.

• Closes Linear regridder causes interpolation errors for masked data in sample data tests #1014
• Link to documentation:

Todo:

• Dataset contains very large unmasked value ESMValTool_sample_data#24 merged
• New release of ESMValTool_sample_data (v0.0.5)
• Update regression data

---

Before you get started

• ☝ Create an issue to discuss what you are going to do

Checklist

• PR has a descriptive title for the changelog
• Labels are assigned so they can be used in the changelog
• Code follows the style guide
• Documentation is available for new functionality
• YAML files pass pre-commit or yamllint checks
• Circle/CI tests pass
• Codacy code quality checks pass
• Documentation builds successfully on readthedocs
• Unit tests are available

If you make backwards incompatible changes to the recipe format:

• Update ESMValTool and link the pull request(s) in the description

---

To help with the number pull requests:

• 🙏 We kindly ask you to review two other open pull requests in this repository

[stefsmeets]

Closed in favour of #1021 which does not touch the sample data repo.

[bouweandela]

Could you please delete the branch once it's not needed anymore? That way we keep the repo clean-ish.