fix(codex): parse jsonl frames linearly

[tarik02]

What Changed

Parse Codex app-server JSONL input by scanning each incoming chunk once and joining pending fragments only when a complete line arrives.

Why

Large thread/resume responses previously rebuilt and rescanned the entire pending line for every chunk, making framing quadratic and causing long disconnect-like stalls.

Checklist

• This PR is small and focused
• I explained what changed and why
• UI changes are not applicable
• Animation and interaction changes are not applicable

Model: GPT-5
Harness: T3 Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 099cdaed54

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".

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	12.1 KiB	12.1 KiB	−3 B (−0.0%)	15.1 KiB	✅
Codex	Thread snapshot wire	5.8 KiB	5.8 KiB	+2 B (+0.0%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.4 KiB	−5 B (−0.1%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.5 KiB	55.5 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	9	9	0 (0.0%)	21	✅
Claude	Total thread wire	12.2 KiB	12.2 KiB	−9 B (−0.1%)	15.1 KiB	✅
Claude	Thread snapshot wire	5.8 KiB	5.8 KiB	−3 B (−0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.4 KiB	6.4 KiB	−6 B (−0.1%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	9	9	0 (0.0%)	21	✅

Baseline: 956252e · PR result: bb4026d · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 84.1 KiB
• Claude decoded thread snapshot: 84.5 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[tarik02]

@codex recheck

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Chef's kiss.

Reviewed commit: bb4026d851

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".