Adapt the ollama streaming tool call

[EntropyYue]

Before submitting your bug report

• I believe this is a bug. I'll try to join the Continue Discord for questions
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS:
- Continue version:
- IDE version:
- Model:
- config:
  

  
  OR link to assistant in Continue hub:

Description

ollama already supports streaming processing, but continue still considers it unsupported, which will prevent continue from streaming back the model's output in agent mode.
And this also causes models using <think>...</think> to fail to output normally: the model's answer will be included within Thought for ....
An alternative is to use an openai-compatible API with the ollama model, which will work normally.

To reproduce

Use the ollama provider with models that support inference and tool calls, such as qwen3

Log output

[jongleeat]

me too on ollama version 0.9.0