feat(rust): add exact-argv streaming (#190)

[konard]

Summary

• add StreamingRunner::from_argv for spawning an executable with exact OS-native arguments
• preserve StreamingRunner::new as the existing shell-command API
• document both invocation modes and add a minor Rust changelog fragment

Root cause and reproduction

StreamingRunner previously stored only a command string and always invoked it through /bin/sh -c or cmd.exe /c. Callers with an executable plus an argument vector therefore had to flatten arguments into shell syntax; the existing POSIX quoting helper is not valid for Windows cmd.exe.

The new integration test invokes the current Rust test binary directly with spaces, shell metacharacters, and quotes in one argument. Before the implementation, it failed to compile because no exact-argv constructor existed. It now verifies successful execution and actual child output.

Verification

• cargo fmt --all -- --check
• cargo clippy --all-targets --all-features
• cargo test --all-features
• cargo test --doc --all-features
• cargo build --release --all-features
• cargo package --allow-dirty
• all Rust release-script unit tests
• changelog fragment validation
• git diff --check

Language parity

This is intentionally Rust-only and labeled parity-exempt. The JavaScript API already supports exact executable/argument invocation through ProcessRunner exec mode; issue #190 identifies the missing Rust equivalent.

Fixes #190

[konard]

Working session summary

Implemented issue #190 and finalized PR #196.

• Added StreamingRunner::from_argv for shell-free, exact OS-native arguments.
• Preserved existing StreamingRunner::new shell behavior.
• Added regression coverage for spaces, quotes, and shell metacharacters.
• Updated Rust documentation and changelog.
• Merged latest main; branch is clean and pushed at cc7d6b7.
• Local checks and final CI pass on Ubuntu, macOS, and Windows.
• PR is mergeable and ready for review.

---

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: $13.312293

📊 Context and tokens usage:

• 269.7K / 200K (135%) input tokens, 35.7K / 128K (28%) output tokens

Total: (269.7K + 9.0M cached) input tokens, 35.7K output tokens, $13.312293 cost

🤖 Models used:

• Tool: OpenAI Codex
• Requested: gpt-5.6-sol
• Thinking level: xhigh (~31999 tokens)
• Model: GPT 5.6 Sol (gpt-5.6-sol)

📎 Log file uploaded as Gist (3628KB)

• 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