fix: sanitize analytics crash and model values

[Optic00]

Summary

• route renderer/child-process crash capture through the existing sanitized exception path
• collapse custom/free-form model names to fixed analytics labels before tracking
• add regression tests for both privacy boundaries

Verification

• npm run test:unit
• npm run typecheck:renderer

---

Summary by cubic

Sanitizes analytics crash reports and model values to prevent leaking local paths or private model names. Fixes curated built-in models being mislabeled as "custom" in analytics.

• Bug Fixes
  • Route renderer/child-process and queued errors through captureSanitizedException so only redacted errors reach PostHog.
  • Expand model allowlist to include all curated built-ins; collapse others to custom. Apply in summarization/transcription contexts and for model_changed events.
  • Add unit tests for crash redaction and to assert every curated public model id passes the allowlist.

^{Written for commit 861cb7c. Summary will update on new commits.}

[cubic-dev-ai]

All reported issues were addressed across 3 files

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[cubic-dev-ai]

1 issue found across 2 files (changes from recent commits).

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="app/analytics-helpers.js">

<violation number="1" location="app/analytics-helpers.js:265">
P3: The new `llama3.2:3b` allowlist entry is not covered by a test assertion. The regression tests verify the other newly added models (`qwen3.5:9b`, `gpt-oss:20b`, and the Bedrock entries), so adding a quick assertion for `llama3.2:3b` would keep the coverage consistent with the test description.</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[cubic-dev-ai]

P3: The new llama3.2:3b allowlist entry is not covered by a test assertion. The regression tests verify the other newly added models (qwen3.5:9b, gpt-oss:20b, and the Bedrock entries), so adding a quick assertion for llama3.2:3b would keep the coverage consistent with the test description.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At app/analytics-helpers.js, line 265:

<comment>The new `llama3.2:3b` allowlist entry is not covered by a test assertion. The regression tests verify the other newly added models (`qwen3.5:9b`, `gpt-oss:20b`, and the Bedrock entries), so adding a quick assertion for `llama3.2:3b` would keep the coverage consistent with the test description.</comment>

<file context>
@@ -255,22 +255,33 @@ function captureSanitizedException(posthogClient, err, distinctId) {
   'gemma4:e2b-nvfp4',
   'gemma4:e4b-nvfp4',
   'gemma4:12b-nvfp4',
+  'llama3.2:3b',
+  'qwen3.5:9b',
+  'gpt-oss:20b',
</file context>