Add support for audio and video input

[jackyzjk]

Add input support for ModelScope audio and video, so that Qwen3's audio or visual large models can be invoked.
API reference: https://bailian.console.aliyun.com/?spm=5176.29597918.J_SEsSjsNv72yRuRFS2VknO.2.191e7b08wdOQzD&tab=api#/api/?type=model&url=2712576

[codecov]

Codecov Report

❌ Patch coverage is 23.80952% with 16 lines in your changes missing coverage. Please review.
✅ Project coverage is 98.87%. Comparing base (8e5611c) to head (c58a2bd).

Files with missing lines	Patch %	Lines
chat.go	20.00%	16 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1083      +/-   ##
==========================================
- Coverage   99.59%   98.87%   -0.72%     
==========================================
  Files          34       34              
  Lines        2206     2226      +20     
==========================================
+ Hits         2197     2201       +4     
- Misses          6       22      +16     
  Partials        3        3              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.