[torch.compile] add a flag to track batchsize statistics

[youkaichao]

I find that #11031 only records the statistics from the engine level, which is not enough.

we need to record the forward batchsize for every model forward, including using cudagraph or not.

then we can have accurate statistics of batchsizes, and can optimize the shapes we want to compile.

[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.

🚀

[mgoin]

This is a great idea, and could be very useful when benchmarking serving benchmarks cc @robertgshaw2-neuralmagic @tlrmchlsmth