Phase 003: remove the coding-agent and ekko-agent runtimes

Every chat session now runs through the Hermes bridge, so the three chat
chains become feature-equivalent instead of three partially-maintained
dispatch paths. Keeping the coding-agent and in-house agent runtimes alive
meant every bridge change had to be mirrored into two more runtimes that
had already drifted.

Historical rows are read, not migrated: sessions stored with the retired
source normalize to cli on load, workflow graphs naming the retired agents
keep loading and run on the bridge, and the visible-source check plus
SOURCE_LABELS still recognize the old value so those sessions stay listed.

The public surface follows the code: the OpenAPI generator and the MCP
chat-run tool schema drop the six retired entry fields so external callers
are not told about runtimes that no longer exist.

harness-check gains a RETIRED_RUNTIME scan so the residual check is a
mechanical gate instead of a manual grep, with the two surviving read-side
compat sites recorded with reasons.

Refs: docs/phases/phase_003.md

tests/server/run-chat-queued-item.test.ts

@@ -2,7 +2,6 @@ import { beforeEach, describe, expect, it, vi } from 'vitest'
 
 const handleBridgeRunMock = vi.hoisted(() => vi.fn(async () => {}))
 const resumeBridgeRunMock = vi.hoisted(() => vi.fn(async () => {}))
-const handleCodingAgentRunMock = vi.hoisted(() => vi.fn(async () => {}))
 const loadSessionStateFromDbMock = vi.hoisted(() => vi.fn())
 const ensureReadyMock = vi.hoisted(() => vi.fn())
 const sessionCommandMocks = vi.hoisted(() => ({
@@ -30,10 +29,6 @@ vi.mock('../../packages/server/src/services/hermes/run-chat/load-state', () => (
   resolveRunSource: vi.fn((source?: string) => source || 'cli'),
 }))
 
-vi.mock('../../packages/server/src/services/hermes/run-chat/handle-coding-agent-run', () => ({
-  handleCodingAgentRun: handleCodingAgentRunMock,
-}))
-
 vi.mock('../../packages/server/src/services/hermes/run-chat/session-command', () => sessionCommandMocks)
 
 vi.mock('../../packages/server/src/services/hermes/agent-bridge', () => ({
@@ -294,7 +289,7 @@ describe('ChatRunSocket queued bridge runs', () => {
     expect(call[6]).toBe(false)
   })
 
-  it('queues coding-agent messages while a coding-agent turn is active', async () => {
+  it('queues messages carrying a retired legacy source while a turn is active', async () => {
     const { ChatRunSocket } = await import('../../packages/server/src/services/hermes/run-chat')
     const { handlers, io, namespace, socket } = makeServerHarness()
     const server = new ChatRunSocket(io as any)
@@ -305,33 +300,31 @@ describe('ChatRunSocket queued bridge runs', () => {
       isAborting: false,
       events: [],
       queue: [],
-      source: 'coding_agent',
+      source: 'retired_source',
     })
 
     await handlers.get('run')?.({
       session_id: 'session-1',
-      input: 'queued codex follow-up',
-      source: 'coding_agent',
-      coding_agent_id: 'codex',
-      queue_id: 'queue-codex',
+      input: 'queued follow-up',
+      source: 'retired_source',
+      queue_id: 'queue-legacy',
       model: 'gpt-5-codex',
       provider: 'openai-codex',
       profile: 'default',
     })
 
-    expect(handleCodingAgentRunMock).not.toHaveBeenCalled()
+    expect(handleBridgeRunMock).not.toHaveBeenCalled()
     expect((server as any).sessionMap.get('session-1').queue).toEqual([
       expect.objectContaining({
-        queue_id: 'queue-codex',
-        input: 'queued codex follow-up',
-        source: 'coding_agent',
-        codingAgentId: 'codex',
+        queue_id: 'queue-legacy',
+        input: 'queued follow-up',
+        source: 'retired_source',
       }),
     ])
     expect(namespace.to).toHaveBeenCalledWith('session:session-1')
   })
 
-  it('dequeues coding-agent messages when an external coding-agent run completes', async () => {
+  it('dequeues messages carrying a retired legacy source on the bridge', async () => {
     const { ChatRunSocket } = await import('../../packages/server/src/services/hermes/run-chat')
     const { io, socket } = makeServerHarness()
     const server = new ChatRunSocket(io as any)
@@ -341,27 +334,25 @@ describe('ChatRunSocket queued bridge runs', () => {
       isAborting: false,
       events: [],
       queue: [{
-        queue_id: 'queue-codex',
-        input: 'queued codex follow-up',
-        source: 'coding_agent',
-        codingAgentId: 'codex',
+        queue_id: 'queue-legacy',
+        input: 'queued follow-up',
+        source: 'retired_source',
         model: 'gpt-5-codex',
         provider: 'openai-codex',
         profile: 'default',
         originSocketId: socket.id,
       }],
-      source: 'coding_agent',
+      source: 'retired_source',
     })
 
     ;(server as any).markExternalRunCompleted('session-1', 'run.completed')
 
-    await vi.waitFor(() => expect(handleCodingAgentRunMock).toHaveBeenCalled())
-    const call = handleCodingAgentRunMock.mock.calls.at(-1)!
+    await vi.waitFor(() => expect(handleBridgeRunMock).toHaveBeenCalled())
+    const call = handleBridgeRunMock.mock.calls.at(-1)!
     expect(call[2]).toEqual(expect.objectContaining({
-      input: 'queued codex follow-up',
-      source: 'coding_agent',
-      coding_agent_id: 'codex',
-      queue_id: 'queue-codex',
+      input: 'queued follow-up',
+      source: 'retired_source',
+      queue_id: 'queue-legacy',
     }))
     expect((server as any).sessionMap.get('session-1').queue).toEqual([])
   })