ttnn.squeeze unable to handle tiled inputs

[yugi957]

Describe the bug
(core.cpp:49: tt::exception\ninfo:\nUnable to reshape a tensor in TILE_LAYOUT to non-tile height and width! Please convert the tensor to ROW_MAJOR_LAYOUT first.)
Squeeze can't handle tiled inputs.

To Reproduce
Steps to reproduce the behavior (make sure environment is built and you have access to elastic cloud/corporate):

1. Go to 'tt_metal/tests/sweep_framework'
2. Run 'python3 parameter_generator.py --elastic <cloud/corporate> --module-name sweeps.data_movement.squeeze.squeeze_pytorch2' to generate tensor parameters
3. Run 'python3 runner.py --elastic <cloud/corporate> --module-name sweeps.data_movement.squeeze.squeeze_pytorch2' to run tests
4. Go to Elastic DB dashboard to view recent sweep run and filter exceptions by the above exception.

Expected behavior
Should run with no errors and output the squeezed tensor exactly as PyTorch does.

Additional context
This was about 35% of total squeeze fails.