fix(search): align global search telemetry with search_analytics schema

[adm01-debug]

Contexto

Follow-up cirúrgico do #329 (já mergeado). O PR #329 corrigiu o mesmo bug em useProductAnalytics.ts, mas passou batido em useGlobalSearch.ts (linha ~713) — que tem o mesmo insert quebrado em search_analytics.

Bug

O insert estava enviando duas colunas que não existem em public.search_analytics:

Código antes	Schema real
seller_id: sellerId	coluna correta é user_id
filters_used: { ... } (jsonb-like)	coluna correta é search_context text

Schema confirmado via SQL contra o banco vivo (`doufsxqlfjyuvxuezpln`):

```
search_analytics columns:
id uuid NOT NULL
user_id uuid NULL
search_term text NOT NULL
results_count integer NOT NULL
search_context text NULL
created_at timestamptz NOT NULL
```

Impacto runtime: todo insert da busca global retornava `PGRST204` (column not found) silenciosamente, quebrando completamente a telemetria de busca. O `.then(() => undefined, () => undefined)` engolia o erro.

Fix

```diff

•      const sellerId = data.user?.id;
  

•      if (!sellerId) return;
  

•      const userId = data.user?.id;
  

•      if (!userId) return;
       return supabase
         .from('search_analytics')
         .insert({
  

•          seller_id: sellerId,
  

•          user_id: userId,
           search_term: searchQuery.toLowerCase().trim().slice(0, 200),
           results_count: finalResults.length,
  

•          filters_used: { latency_ms: latencyMs, intent_type: intent.type },
  

•          search_context: JSON.stringify({ latency_ms: latencyMs, intent_type: intent.type }),
         })
  

```

O payload (`latency_ms` + `intent_type`) é serializado como JSON dentro de `search_context` porque a coluna é `text`, não `jsonb`. Quem for consumir essa telemetria depois pode fazer `(search_context::jsonb)->>'intent_type'` no SQL.

Validação

• ✅ 4 inserções / 4 deleções (mesmo arquivo, mesmo bloco)
• ✅ ESLint baseline gate no pre-push: zero regressões (drift positivo: −8 erros)
• ⏳ Typecheck completo + build deixados para o CI (gateway timeout no run local com 3600 arquivos)
• ⚠️ Não foi possível exercitar a UI ao vivo neste ambiente

Por que não foi no #329

O #329 tocou `useProductAnalytics.ts` (telemetria de visualização de produto) mas não `useGlobalSearch.ts` (telemetria de busca global). Mesma tabela, mesma raiz, dois call sites — o segundo foi descoberto na auditoria pós-merge.

Próximo (não escopo deste PR)

A descrição do #329 menciona drift de schema na personalização (`product_group_` / `product_component_` referenciando colunas que não existem no DB). Continua aberto e requer decisão de produto.

---

Summary by cubic

Fixes global search telemetry to match the public.search_analytics schema, preventing PGRST204 column errors and restoring metrics collection. Applies the same fix missed in useGlobalSearch.ts that was done in #329.

• Bug Fixes
  • Use user_id instead of seller_id.
  • Write payload to search_context (text) as a JSON string: { latency_ms, intent_type }.
  • Keep early return if no userId to avoid invalid inserts.

^{Written for commit 07e9b9f. Summary will update on new commits. Review in cubic}

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[coderabbitai]

Warning

Review limit reached

@adm01-debug, we couldn't start this review because you've used your available PR reviews for now.

Your plan includes 5 reviews of capacity. Refill in 22 minutes and 51 seconds.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

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

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than trial, open-source, and free plans. In all cases, review capacity refills continuously over time.

Please see our FAQ for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 1fd1e5f9-0735-4a49-93d3-743b2c87bc12

📥 Commits

Reviewing files that changed from the base of the PR and between 0277526 and 07e9b9f.

📒 Files selected for processing (1)

• src/components/search/useGlobalSearch.ts

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/search-analytics-global-search-2026-05-25

---

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

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
we-dream-big	Ready	Preview, Comment	May 25, 2026 1:38pm

[supabase]

This pull request has been ignored for the connected project doufsxqlfjyuvxuezpln because there are no changes detected in supabase directory. You can change this behaviour in Project Integrations Settings ↗︎.

---

Preview Branches by Supabase.
Learn more about Supabase Branching ↗︎.

[Copilot]

Pull request overview

This PR restores global search telemetry by aligning the search_analytics insert in useGlobalSearch.ts with the live public.search_analytics schema (matching the fix previously applied elsewhere in #329), preventing PGRST204 “column not found” failures and allowing metrics to be recorded again.

Changes:

• Writes user_id instead of the non-existent seller_id.
• Serializes the telemetry payload into search_context (text) instead of the non-existent filters_used.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.