Remove obsolete parameter mapping between pytorch and keras

The pytorch parser still uses a mapping between tensor names from keras to pytorch which has become obsolete. In its current form, it breaks BatchNorm layers. This PR simply removes this functionality, restoring BatchNorm functionality and removing clutter. A test for BatchNorm in pytorch is added to prevent further breaking of this functionality

Type of change

For a new feature or function, please create an issue first to discuss it with us before submitting a pull request.

Note: Please delete options that are not relevant.

[x] Bug fix (non-breaking change that fixes an issue)

Tests

Ran pytests to verify that nothing gets broken, verified that BatchNorm layers parse successfully after the fix.

Checklist

[x] I have read the guidelines for contributing.
[x] My changes generate no new warnings.
[x] I have installed and run pre-commit on the files I edited or added.
[x] I have added tests that prove my fix is effective or that my feature works.

fastmachinelearning / hls4ml