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

bin/hermes-ui-mcp.mjs

@@ -856,13 +856,13 @@ const tools = [
         },
         source: {
           type: 'string',
-          enum: ['cli', 'coding_agent', 'global_agent'],
-          description: 'Optional run backend source.',
+          enum: ['cli', 'global_agent', 'workflow'],
+          description: 'Run backend source. Use cli for Hermes bridge runs or global_agent for global-agent sessions.',
         },
         session_source: {
           type: 'string',
-          enum: ['global_agent'],
-          description: 'Optional session source marker.',
+          enum: ['global_agent', 'workflow'],
+          description: 'Marks the session as launched from the global agent or from a workflow run.',
         },
         instructions: {
           type: 'string',
@@ -876,34 +876,6 @@ const tools = [
           type: 'string',
           description: 'Optional per-run reasoning effort override.',
         },
-        coding_agent_id: {
-          type: 'string',
-          enum: ['claude-code', 'codex'],
-          description: 'Coding agent id when source is coding_agent.',
-        },
-        agent_id: {
-          type: 'string',
-          enum: ['claude-code', 'codex'],
-          description: 'Alias for coding_agent_id.',
-        },
-        mode: {
-          type: 'string',
-          enum: ['scoped', 'global'],
-          description: 'Coding-agent launch mode.',
-        },
-        baseUrl: {
-          type: 'string',
-          description: 'Optional provider base URL for coding-agent runs.',
-        },
-        apiKey: {
-          type: 'string',
-          description: 'Optional provider API key for coding-agent runs.',
-        },
-        apiMode: {
-          type: 'string',
-          enum: ['chat_completions', 'codex_responses', 'anthropic_messages'],
-          description: 'Optional provider wire API mode for coding-agent runs.',
-        },
         timeout_ms: {
           type: 'number',
           description: 'Maximum time to wait for a terminal run event.',
@@ -1505,12 +1477,6 @@ async function callTool(name, args = {}) {
           'instructions',
           'workspace',
           'reasoning_effort',
-          'coding_agent_id',
-          'agent_id',
-          'mode',
-          'baseUrl',
-          'apiKey',
-          'apiMode',
           'timeout_ms',
           'include_events',
         ]),