Fix stream-json tool use details

[konard]

Fixes #281

Summary

• Read tool names from the runtime tool part's tool field.
• Read tool arguments from the tool state's input field.
• Replace the obsolete synthetic unit fixture with a schema-accurate write tool event.
• Add a patch changeset for the JavaScript package.

Reproduction and coverage

The regression fixture mirrors the real MessageV2.ToolPart emitted for a streamed tool call:

{
  type: 'tool',
  tool: 'write',
  state: {
    status: 'pending',
    input: { filePath: 'hi.txt', content: 'hi\n' },
  },
}

Before the fix, the converter produced name: "unknown" and input: {} because it looked for the obsolete state.tool.name and state.tool.parameters fields. The updated test asserts the emitted event contains name: "write" and the original input.

Verification

cd js
bun test ./tests/json-standard-unit.js
npm run check
npm test

Observed locally: focused converter tests passed (18/18), repository checks passed, and the full JavaScript suite passed with 609 pass / 4 todo / 0 fail.

[konard]

Working session summary

Implemented and finalized the fix in PR #282.

• Corrected stream-json tool names and inputs using the actual runtime schema.
• Added a reproducing regression test.
• Added a JavaScript patch changeset.
• Local suite: 609 passed, 4 existing todos, 0 failed.
• Lint, formatting, and file-size checks passed.
• CI passed on Ubuntu, macOS, and Windows.
• PR is ready for review and cleanly mergeable.
• Commit: ee63864 (Fix stream-json tool use details)

---

This summary was automatically extracted from the AI working session output.

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Model: GPT-5.6 Sol
• Provider: OpenAI
• Public pricing estimate: $3.858589

📊 Context and tokens usage:

• 110.8K / 200K (55%) input tokens, 6.8K / 128K (5%) output tokens

Total: (110.8K + 2.4M cached) input tokens, 6.8K output tokens, $3.858589 cost

🤖 Models used:

• Tool: OpenAI Codex
• Requested: gpt-5.6-sol
• Thinking level: low (~7999 tokens)
• Model: GPT 5.6 Sol (gpt-5.6-sol)

📎 Log file uploaded as Gist (1223KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

🎉 Auto-merged

This pull request has been automatically merged by hive-mind.

• All CI checks have passed

---

Auto-merged by hive-mind with --auto-merge flag