Unify admin storage usage metrics

[Rabithua]

Summary

• use the authoritative resource storage ledger for the official rote.ink admin storage ranking
• keep the existing attachment-size calculation for self-hosted instances
• use the same metric for displayed usage and ranking order

Root cause

The user resource screen read resource_storage_accounts.used_bytes, while the admin dashboard still summed client-declared attachment sizes. This produced different totals for the same user.

Validation

• bun run build
• bunx eslint utils/dbMethods/admin.ts
• git diff --check

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ebf1b1ec6d

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".

[chatgpt-codex-connector]

Keep unreconciled storage usage unknown

When billing is enabled while the initial storage reconciliation is pending or failed, this query converts a missing account to zero and exposes nonzero accounts even if reconciledAt is null. In contrast, reportedOfficialUsedBytes in resources/service.ts returns null until the account or global reconciliation is complete. Consequently, the admin dashboard can filter out legacy users as having no usage or rank them using a partial ledger, so it still disagrees with the user resource screen during rollout; preserve the unreconciled state instead of silently coercing it to a usable total.

AGENTS.md reference: AGENTS.md:L48-L50

Useful? React with 👍 / 👎.