feat(backend): ingest GitHub issues with idempotent upsert (#74)

[Cedarich]

🚀 PR: Ingest GitHub Issues into Backend Database (#74)

📌 Summary

• Adds an ingestion path that fetches GitHub issues for specified repositories and persists them with idempotent upsert semantics.
• Lays the groundwork for reward processing and other downstream features.

---

🔍 Scope

• Endpoint: POST /admin/github/sync
• Accepts payload:

  {
    "repos": ["org/repo", ...],
    "since": "2024-01-01T00:00:00Z" // optional
  }
  
  

close #74

[oscarwroche]

The API interaction needs to be handled in the infrastructure layer instead of the application layer. Please create a service trait in the domain layer and create an implementation in the infrastructure. The application side should have no direct references to infrastructure (Github API, Postgres, etc) concerns

[Cedarich]

@oscarwroche Please review again This refactor enforces clean architecture boundaries: application depends on the GithubApiService trait, while infrastructure provides the HTTP implementation. No direct reqwest or Postgres references remain in the application layer.

[oscarwroche]

@Cedarich Looks great overall! Just a comment about removing the query! macro in the indexer

[Cedarich]

@oscarwroche

• I restored compile-time checking in the indexer by switching the list query to sqlx::query! and kept the event_type JSON handling intact.
• The offline cache for the SELECT is already generated and committed at indexer/.sqlx/query-c8d8e35…json .
• The insert_many path is temporarily using sqlx::query (dynamic) to unblock diagnostics while my local database is unavailable. This means compile-time validation for the insert is pending.

[oscarwroche]

I still don't know why there are changes to the indexer files, they don't have anything to do with the goal of this PR. Creating this migration is unnecessary

[Cedarich]

Please review 🙏

[oscarwroche]

why is the migration renamed ?

[oscarwroche]

And I still don't understand why there are any changes to the indexer files

[oscarwroche]

why is the migration renamed ?

[oscarwroche]

And I still don't understand why there are any changes to the indexer files

[joelamouche]

@Cedarich what's the update on this pls?

[joelamouche]

close due to inactivity