Deduplicate Codex usage across Pi forked sessions

[arisgysel-design]

Summary

• count inherited Pi messages only once across parent and fork session files
• keep new fork responses and unrelated short-ID collisions distinct
• derive active sessions from Pi's parentSession lineage so scan order cannot change session totals

Testing

• bash test/shell.d/agent-usage-codex-scanner-test.sh
• python3 -m py_compile bin/omarchy-agent-usage-codex
• bash bin/omarchy commands --check (437 commands)

Fixes #8564

[Copilot]

Pull request overview

Deduplicates inherited Codex usage across Pi/OMP fork lineages while preserving usage from new responses and unrelated ID collisions.

Tip

If you aren't ready for review, convert to a draft PR.
Click "Convert to draft" or run gh pr ready --undo.
Click "Ready for review" or run gh pr ready to reengage.

Changes:

• Uses message identity and parentSession ancestry for deterministic usage attribution.
• Expands scanner tests for forks, scan ordering, and ID collisions.

Reviewed changes

Copilot reviewed 1 out of 2 changed files in this pull request and generated no comments.

File	Description
bin/omarchy-agent-usage-codex	Deduplicates inherited messages and recalculates session ownership.
test/shell.d/agent-usage-codex-scanner-test.sh	Tests fork accounting and collision handling.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Brams-s]

Related native-collector correction: #10531 skips repeated token_count notifications carrying an unchanged cumulative snapshot. I reproduced one 110-token request becoming 220 tokens in current quattro.

It stays within scan_native_codex_sessions() plus cache invalidation and tests; Pi lineage/deduplication remains with this PR. I checked that your current patch applies cleanly alongside it. Linking them so the two distinct duplication cases can be reviewed together without combining their implementations.