[Feature][WIP] Basic version of supporting streaming output

[Gaohan123]

Purpose

This PR supports basic straming output for both offline and online. For text, it can output token by token or fixed chunk size. For audio output, it depends on later feature of chunked async pipeline across multiple stages.

Test Plan

Test Result

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft.

BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)

[hu-sp]

@Gaohan123 @hsliuustc0106 Will the 1230 release #165 include support for returning audio data when stream: true is enabled #373 ?

[wonjerry]

This feature is really needed! Even just streaming for text output would be great.

[hsliuustc0106]

for this feature, do we support text streaming for the first stage and leave the audio streaming for the follow-up PR?