Remove uses of get_observer

[kylesayrs]

Purpose

• Be consistent about how the observer of the quantization arguments is referenced
  From compressed tensors:

def get_observer(self):
    return self.observer

Postrequisites

• Depreciate get_observer neuralmagic/compressed-tensors#214

Changes

• Remove all uses of quantization_args.get_observer()

Testing

grep -r '\.get_observer()' src examples tests

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

[horheynm]

This is all the pathway that uses get_observer?

[kylesayrs]

@horheynm

Testing

grep -r '\.get_observer()' src examples tests

[dsikka]

lets rebase/make sure tests pass