fix(proxy): capture streaming token usage for Ollama providers

[SebConejo]

Summary

• Inject stream_options.include_usage: true for ollama and ollama-cloud endpoints in streaming requests, matching the existing behavior for openai and openrouter
• Without this flag, Ollama servers omit token usage data from the final SSE chunk, resulting in 0 tokens and $0.00 costs in the dashboard
• Add test coverage for both Ollama variants

Closes #1585

---

Summary by cubic

Fix missing token usage in streaming responses from ollama and ollama-cloud by injecting stream_options.include_usage, matching openai/openrouter. Restores accurate token counts and costs in the dashboard. Closes #1585.

• Bug Fixes
  • Inject stream_options.include_usage: true for streaming requests to ollama and ollama-cloud.
  • Added tests to assert the injection for both providers.

^{Written for commit aa004e3. Summary will update on new commits.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.92%. Comparing base (67f8852) to head (aa004e3).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1592   +/-   ##
=======================================
  Coverage   97.92%   97.92%           
=======================================
  Files         118      118           
  Lines        8675     8675           
  Branches     3265     3265           
=======================================
  Hits         8495     8495           
  Misses        178      178           
  Partials        2        2           

Flag	Coverage Δ
frontend	97.91% <ø> (ø)
shared	100.00% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

🚀 New features to boost your workflow:

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

[cubic-dev-ai]

No issues found across 3 files