fix(flows): preserve literal tool-call markup in tool-less completions

[senamakel]

Summary

• Preserves literal <tool_call>...</tool_call> examples in one-shot flow completions when no tools were advertised.
• Keeps prompt-guided tool-call extraction unchanged for requests that do advertise tools, in both buffered and streaming paths.
• Removes the now-unused compatibility helpers from the original branch after current main superseded the ChatModel migration in feat(inference): Phase 3 P3-B — crate-native turn path + delete compatible*.rs [WIP] #4784.

Problem

After the original ChatModel migration, ProviderModel parsed tool-call markup from every text response. A tool-less flow prompt that legitimately requested a literal tool-call example therefore lost that text and surfaced a synthetic tool call instead.

Solution

Text fallback parsing is now gated by whether ModelRequest.tools is non-empty. Native provider tool calls still take precedence. Regression tests cover both tool-less literal preservation and tool-enabled extraction.

Submission Checklist

• Tests added or updated — focused model-response tests cover both branches; tinyflows capability tests remain green.
• Diff coverage ≥ 80% — new branch behavior is covered directly.
• Coverage matrix updated — N/A: internal inference adapter bug fix with co-located unit coverage.
• All affected feature IDs — N/A: no user-facing feature catalog change.
• No new external network dependencies introduced.
• Manual smoke checklist — N/A: deterministic adapter behavior is unit-tested.
• Linked issue — review thread on PR fix(flows): preserve literal tool-call markup in tool-less completions #4750; original migration tracked by Complete tinyagents inference migration (Motion B): crate-native provider clients, delete duplicated provider stack #4727 / Improve agent harness with LangGraph-style state machine or Polaris-like architecture #4249.

Validation

• cargo fmt --manifest-path Cargo.toml -- --check
• cargo check --manifest-path Cargo.toml
• cargo check --manifest-path app/src-tauri/Cargo.toml
• cargo test --manifest-path Cargo.toml --lib openhuman::tinyagents::model::g1_usage_tests
• cargo test --manifest-path Cargo.toml --lib openhuman::tinyflows::caps::tests
• Pre-push: Prettier, ESLint, TypeScript compile, Tauri Rust check

The full unfiltered Rust suite is currently blocked by a pre-existing deterministic stack overflow in openhuman::agent::harness::session::tests::turn_dispatches_spawn_subagent_through_full_path; the exact test fails identically when rerun alone.

Impact

• Platforms: core behavior used by desktop/web/CLI flow execution.
• Behavior: tool-less completions retain literal markup; tool-enabled completions continue extracting calls.
• Compatibility: no persistence or RPC schema changes.

[coderabbitai]

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 37 seconds

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: b13fd43e-37c8-4ab2-8b9c-9364b46d4503

📥 Commits

Reviewing files that changed from the base of the PR and between e0a7b66 and dd2dee7.

📒 Files selected for processing (1)

• src/openhuman/tinyagents/model.rs

---

_{Comment @coderabbitai help to get the list of available commands.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 16ae696494

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".