feat(issues): add Linear as a standalone issue tracker

[juliusmarminge]

Problem

Linear and Jira track issues without hosting repositories, so the git-host trackers in #19259–#19261 can't reach them.

Change

• Standalone trackers: StandaloneIssueTrackerApi in issue-tracker-core adds a descriptor (display name, host, scope noun such as "team" or "project", credential label) and probeCredential. The server stores one token per tracker in the secret store and checks it before saving.
• Project binding: a project can bind one scope of a standalone tracker (migration 063, issue_tracker_bindings). The bound tracker sits beside the git host's, so a GitHub repository planned in Linear lists both, and both are writable. A project without a remote can use a standalone tracker alone. Disconnecting a tracker removes its bindings.
• Linear (packages/issue-tracker-linear, GraphQL over HttpClient): keys such as ENG-123, scoped to a team. Workflow states keep their names; completed and canceled count as closed and double as close reasons. Plain templates open as drafts; form and rich templates link out to Linear, since its API can't submit them.
• Settings → Source control → Issue trackers: connect or disconnect each standalone tracker the server reports, and bind or unbind projects. The UI iterates trackers generically.

Adding Jira takes a packages/issue-tracker-jira driver, one row in builtInIssueTrackers.ts, and an optional settings icon. Contracts, RPCs, services, and the settings UI need no change.

Verification

Linear tracker fixture tests, the issue service binding test, and RPC and client permission tests pass. Typecheck is clean for contracts, both tracker packages, client-runtime, server, and web. Untested against a live Linear workspace.

Rebuilt from #6315 by @Bil0000, which this stack uses as its reference for host APIs and UX.

Implemented with Claude Opus 5.5 and Claude Fable 5.1 in T3 Code, using the Claude Code harness.

🤖 Generated with Claude Code

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

ℹ️ No successful main baseline artifact is available yet. This run establishes the initial measurement.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	—	4.9 KiB	—	6.8 KiB	✅
Codex	Thread snapshot wire	—	3.8 KiB	—	4.9 KiB	✅
Codex	Live turn WebSocket wire	—	1.2 KiB	—	2.0 KiB	✅
Codex	Live turn WebSocket decoded	—	20.8 KiB	—	29.3 KiB	✅
Codex	Live turn messages	—	1	—	8	✅
Claude	Total thread wire	—	5.0 KiB	—	6.8 KiB	✅
Claude	Thread snapshot wire	—	3.8 KiB	—	4.9 KiB	✅
Claude	Live turn WebSocket wire	—	1.2 KiB	—	2.0 KiB	✅
Claude	Live turn WebSocket decoded	—	21.2 KiB	—	29.3 KiB	✅
Claude	Live turn messages	—	2	—	8	✅

Baseline: unavailable · PR result: 322107b · 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: 108.5 KiB
• Claude decoded thread snapshot: 108.8 KiB

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