[V1][Bugfix] Propagate V1 LLMEngine properly

[comaniac]

#9888 makes V1 LLMEngine import lazily, but failed to pass the right engine to the constructor, resulting in crashing when launching the engine.

We should really make some basic tests for v1 ASAP...

cc @joerunde @WoosukKwon @youkaichao

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

[njhill]

@comaniac I'm just checking whether this makes sense given that we won't use MQLLMEngine with V1.

[njhill]

Yes this won't be needed once we merge #9826 (hopefully imminently), which reverts the use of v1 engine in mutliprocessing/engine.py.

[comaniac]

Ok make sense. Than we could close this one once #9825 is merged.

[joerunde]

Ah, good catch @comaniac!

I will start enabling more tests with v1, but won't do any of the engine/multiprocessing ones since this won't be used with v1

[DarkLight1337]

Closing as #9826 has been merged.