fix: provide fallback context window values for Ollama and LM Studio models

[roomote]

This PR attempts to address Issue #7674. Feedback and guidance are welcome.

Problem

The task header was displaying incorrect max context window for Ollama (showing "used/1" instead of actual max tokens). The issue was that info was always undefined at line 257 in useSelectedModel.ts when routerModels.ollama was an empty object or the specific model wasn't found.

Solution

• Add fallback ModelInfo when routerModels.ollama or lmStudioModels return undefined
• Provide reasonable default values (8192 for context window and max tokens)
• Apply the same fix to both Ollama and LM Studio providers for consistency

Changes

• Modified webview-ui/src/components/ui/hooks/useSelectedModel.ts to provide fallback values when model info is undefined
• Set appropriate support flags (Ollama supports prompt cache, LM Studio doesn't)

Testing

• All existing tests pass
• Linting passes
• Type checking passes

Fixes #7674

---

Important

Fixes incorrect context window display for Ollama and LM Studio models by providing fallback values in useSelectedModel.ts.

• Behavior:
  • Fixes incorrect context window display for Ollama and LM Studio models in useSelectedModel.ts by providing fallback values when model info is undefined.
  • Sets default values of 8192 for both context window and max tokens.
  • Ensures Ollama supports prompt cache, while LM Studio does not.
• Testing:
  • All existing tests pass.
  • Linting and type checking pass.

^{This description was created by for 5b55f6e. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Reviewing my own code is like debugging in a mirror - everything looks backward but the bugs are still mine.

[roomote]

The fallback logic here is duplicated with the LM Studio case below. Could we extract this into a shared helper function to reduce duplication? Something like:

[roomote]

Is 8192 the right default for all Ollama models? Some models support much larger context windows. Could we consider making this configurable or perhaps use a more generous default like 32768?

[roomote]

The non-null assertion here could be avoided with better type checking. Consider:

[roomote]

Should we add test coverage for these fallback scenarios? I noticed there are no tests for Ollama or LM Studio providers in the test file. This would help ensure the fallback behavior works correctly and prevent regressions.

[daniel-lxs]

Closing in favor of #7679 which provides a more comprehensive solution by fixing the root cause rather than using hardcoded fallback values.