Add Claude stream-json stdin support

[konard]

Fixes #268

Summary

• Add --input-format stream-json for Claude-compatible JSONL stdin frames.
• Parse user/user_prompt, system, and interrupt frames one line at a time, with auto-merge disabled in stream-json mode.
• Emit Claude-style replay acknowledgements for consumed stream-json user prompts when --output-format stream-json is used.
• Add --output-format stream-json as an alias for Claude NDJSON output, plus Rust CLI option parity and release fragments.

Reproduction and Verification

• Reproduces the issue at the queue boundary: stream-json frames are no longer merged and each user frame becomes a separate prompt.
• Added js/tests/input-queue.js coverage for Claude user, simplified user_prompt, system, interrupt, and invalid non-text frames.

Tests

• cd js && bun test ./tests/input-queue.js ./tests/cli.ts ./tests/cli_options.ts
• cd rust && cargo test --test cli --test cli_options
• cd js && npm test
• cd js && npm run check
• cd rust && cargo test
• git diff --check

[konard]

Working session summary

Implemented and pushed the fix to PR #269: #269

What changed:

• Added --input-format stream-json for Claude-compatible JSONL stdin frames.
• Added --output-format stream-json alias mapping to Claude NDJSON output.
• Parses user/user_prompt, system, and interrupt frames without auto-merging.
• Emits Claude-style replay acknowledgements for consumed stream-json user prompts.
• Added JS parser/CLI tests, Rust CLI option parity, and release fragments.

Verification:

• cd js && npm test
• cd js && npm run check
• cd rust && cargo test
• git diff --check
• Latest JS and Rust CI runs on commit 3627542 are passing.

PR #269 is marked ready for review, and the working tree is clean.

---

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-5.5
• Provider: OpenAI
• Public pricing estimate: $19.720062

📊 Context and tokens usage:

• peak request: 15.0M / 1.1M (1431%) input tokens, 42.6K / 128K (33%) output tokens

Total: (309.0K + 14.7M cached) input tokens, 42.6K output tokens, $19.720062 cost

🤖 Models used:

• Tool: OpenAI Codex
• Requested: gpt-5.5
• Model: GPT-5.5 (gpt-5.5)

📎 Log file uploaded as Repository (40675KB)

• View complete solution draft log

---

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

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag