fix: fix Dockerfile ubi9 base

[YashasviChaurasia]

Description of the change

Default to use : ubi9/ubi:9.6-1758184894 in Dockerfile
https://catalog.redhat.com/en/software/containers/ubi9/ubi/615bcf606feffc5384e8452e?image=68cc233126a59a51953ccb56#get-this-image

The latest base image: ubi9/ubi:9.6-1760340943 has issue while building image test which we are having trouble to replicate on vm as the same Dockerfile builds fine without any error.

> [python-installations  8/14] RUN --mount=type=cache,target=/home/tuning/.cache/pip,uid=1000     python -m pip install --user wheel &&     python -m pip install --user "$(head bdist_name)" &&     python -m pip install --user "$(head bdist_name)[flash-attn]" &&     python -m pip install --user --no-build-isolation "$(head bdist_name)[mamba]":
#33 442.7         File "/tmp/pip-build-env-zejlth0s/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 512, in run_setup
#33 442.7           super().run_setup(setup_script=setup_script)
#33 442.7         File "/tmp/pip-build-env-zejlth0s/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 317, in run_setup
#33 442.7           exec(code, locals())
#33 442.7         File "<string>", line 22, in <module>
#33 442.7       ModuleNotFoundError: No module named 'torch'
#33 442.7       [end of output]
#33 442.7   
#33 442.7   note: This error originates from a subprocess, and is likely not a problem with pip.
#33 442.7 ERROR: Failed to build 'flash-attn' when getting requirements to build wheel

Related issue number

How to verify the PR

Was the PR tested

• I have added >=1 unit test(s) for every new method I have added.
• I have ensured all unit tests pass

[github-actions]

Thanks for making a pull request! 😃
One of the maintainers will review and advise on the next steps.

[YashasviChaurasia]

The suspected cause for Image test to fail was updated ubi9 base image but it seems to be the issue is independent of the base image..
As the issue persists even after the revert to prev Dockerfile with older base image.

The suspect was due to the fact that all the PRs after PR#617 had the build image failure issue and upon inspecting the build details, the logs pointed towards different base images being used after the above mentioned PR..