[Bugfix][V1] Fix flashinfer sampling

[DefTruth]

small fix base on #14788
Currently, we have errors when using FlashInfer>=v0.2.3 for top-p & top-k sampling. As a workaround, we disable FlashInfer for top-p & top-k sampling by default while FlashInfer>=v0.2.3. The sampling API removes the success return value of all sampling API, which is not compatible with earlier design. reference: https://github.com/flashinfer-ai/flashinfer/releases/tag/v0.2.3

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

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 either: Add ready label to the PR or enable auto-merge.

🚀

[WoosukKwon]

LGTM! Thanks for the PR!

[yzh119]

Hi @WoosukKwon @DefTruth , it should be easy to switch to the new API, maybe @xslingcn can help create a new PR for the new API?

[WoosukKwon]

@yzh119 Thanks for brining this up. We do want to upgrade the version, but didn't have bandwidth recently. We'd really appreciate it if @xslingcn or anyone could take this!