[Test Fix] Add Quantization then finetune tests

[horheynm]

Contingent on merge of huggingface/transformers#34719
^ has been merged not yet released

SUMMARY:
Add test to

• Given a model, oneshot quantize, then run ptq - training.
  Model must be run_compressed = False to run

Note:

• When running finetune on an already optimized (one-shotted) mode, the model needs to be decompressed explicitly using CompressedTensorsConfig. See https://github.com/vllm-project/llm-compressor/pull/964/files#diff-e480ed475c0a5b2beb4052c1dd2aca671999634ace41a5ea017fdff1ce68be0bR130-R135
• Tests using x2 H100s passed

Also fix a bug where in log_sparsification, the layer name is not being recognized so fails. Here nothting is being sparsified, so num params is set to zero

TEST PLAN:
ran the test using transformers main
must pass tests/llmcompressor/transformers/finetune/test_oneshot_then_finetune.py

[github-actions]

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

[horheynm]

/ready

[dsikka]

LGTM.

As per offline conversation, lets decompress before running finetune
and verify if we still need the skipif

[dsikka]

LGTM. please fix quality

[rahul-tuli]

LGTM

[kylesayrs]

Need some clarification on params_quantized, otherwise LGTM!

[kylesayrs]

if getattr(layer, "weight", None) is not None:
    num_params += torch.numel(layer.weight)
if getattr(layer, "bias", None) is not None:
    num_params += torch.numel(layer.bias)