[Android Maintenance] Fix reasoning_effort enum for Duck.ai prompt submitted pixels

[daxmobile]

Task/Issue URL: https://app.asana.com/1/137249556945/project/1213746476312668/task/1215891186389639

Description

Live pixel validation reported that m_aichat_unified_input_prompt_submitted_count and m_aichat_unified_input_prompt_submitted_daily were failing with /reasoning_effort must be equal to one of the allowed values.

Root cause: the reasoning_effort parameter on these two pixels is populated by NativeInputModeWidgetViewModel.fireSubmissionPixels() → getResolvedReasoningEffort(), which returns ReasoningEffort.rawValue — one of none, minimal, low, medium, high (the same resolved effort value that is sent to the backend via sendPrompt). However, the pixel definition declared the enum as ["fast", "reasoning", "extended_reasoning"], which are the ReasoningMode raw values. Those mode values are correctly used by the separate m_aichat_unified_input_reasoning_effort_selected pixel's effort_level parameter (ReasoningMode.toEffortParam()), but not by the prompt-submitted pixels.

This change aligns the reasoning_effort enum on both prompt-submitted pixel definitions with the actual wire values (ReasoningEffort raw values). No code behaviour changes — only the definition in PixelDefinitions/pixels/definitions/duck_chat.json5 is updated. The unrelated reasoning_effort_selected pixel's effort_level enum is intentionally left untouched.

🤖 Android Maintenance Worker

Steps to test this PR

Lint / formatting

• cd PixelDefinitions && npm run lint (prettier — passes locally)
• cd PixelDefinitions && npm run validate-defs-without-formatting (pixel-schema validation)

Note: only a PixelDefinitions JSON5 file changed (no Kotlin/Android source), so ./gradlew spotlessCheck / lint_check have nothing applicable to check for this change. npm run lint (prettier) passed locally; npm run validate-defs-without-formatting could not complete in the sandbox because it fetches the live target app version over the network (ERROR in product.json target version: fetch failed), which the CI sandbox firewall blocks — this is unrelated to the change and will run on CI.

UI changes

Before	After
No UI changes	No UI changes

Note

🔒 Integrity filter blocked 3 items

The following items were blocked because they don't meet the GitHub integrity level.

• #8942 list_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #8860 list_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #8844 list_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by Android Maintenance Worker · ◷

---

Note

Low Risk
Schema-only change in PixelDefinitions with no Android runtime behavior changes; it restores validation for analytics events already emitted by the app.

Overview
Fixes live pixel validation failures on m_aichat_unified_input_prompt_submitted_count and m_aichat_unified_input_prompt_submitted_daily by updating the reasoning_effort parameter schema in duck_chat.json5.

The allowed values change from fast / reasoning / extended_reasoning (UI ReasoningMode) to none / minimal / low / medium / high, matching what the app already sends via getResolvedReasoningEffort() (ReasoningEffort raw values). The parameter description now says resolved reasoning effort. m_aichat_unified_input_reasoning_effort_selected and its effort_level enum are unchanged.

^{Reviewed by Cursor Bugbot for commit e704fae. Bugbot is set up for automated code reviews on this repo. Configure here.}