[Fix] Zca inverse transform fix

[ceroytres]

Description

I noticed that I forgot to include the regularization (epsilon) in computing the inverse ZCA transform. I updated the test to include different epsilon values.

Status

Ready

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• Breaking change (fix or new feature that would cause existing functionality to change)
• New tests added to cover the changes
• Docstrings/Documentation updated

PR Checklist

PR Implementer

This is a small checklist for the implementation details of this PR.

If there are any questions regarding code style or other conventions check out our
summary.

• Did you discuss the functionality or any breaking changes before ?
• Pass all tests: did you test in local ? make test
• Unittests: did you add tests for your new functionality ?
• Documentations: did you build documentation ? make build-docs
• Implementation: is your code well commented and follow conventions ? make lint
• Docstrings & Typing: has your code documentation and typing ? make mypy
• Update notebooks & documentation if necessary

KorniaTeam

KorniaTeam workflow

• Assign correct label
• Assign PR to a reviewer
• Does this PR close an Issue? (add closes #IssueNumber at the bottom if
  not already in description)

Reviewer

Reviewer workflow

• Do all tests pass? (Unittests, Typing, Linting, Documentation, Environment)
• Does the implementation follow kornia design conventions?
• Is the documentation complete enough ?
• Are the tests covering simple and corner cases ?

[edgarriba]

LGTM