fix: qkvsync bug fix

[chichun-charlie-liu]

Description of the change

qkvsync is a function in model_analyzer trying to find Linear layers with common input, e.g. Q, K, V, or gate/up (in Llama arch). The results by this searching, a dict like {"q":"q", "k":"q", "v":"q", ...}, will be saved in qcfg["qkvsync_my_1st_siblings"]. This dict is supposed to contain the full layer name, however, when graph breaks happens, model_analyzer can only identify partial names instead. Later functions attempt to use model.get_submodule() based on this dict will fail.

How to verify the PR

run dq with bamba (ibm-ai-platform/Bamba-9B-V2)

Was the PR tested

• I have added >=1 unit test(s) for every new method I have added (if that coverage is difficult, please briefly explain the reason)
• I have ensured all unit tests pass

Checklist for passing CI/CD:

• All commits are signed showing "Signed-off-by: Name <email@domain.com>" with git commit -signoff or equivalent
• PR title and commit messages adhere to Conventional Commits
• Contribution is formatted with tox -e fix
• Contribution passes linting with tox -e lint
• Contribution passes spellcheck with tox -e spellcheck
• Contribution passes all unit tests with tox -e unit

Note: CI/CD performs unit tests on multiple versions of Python from a fresh install. There may be differences with your local environment and the test environment.