feat(hephaestus): upgrade default model from gpt-5.3-codex to gpt-5.4

Hephaestus now uses gpt-5.4 as its default model across all providers
(openai, github-copilot, venice, opencode), matching Sisyphus's GPT 5.4
support. The separate gpt-5.3-codex → github-copilot fallback entry is
removed since gpt-5.4 is available on all required providers.

src/agents/utils.test.ts

@@ -642,7 +642,7 @@ describe("createBuiltinAgents with requiresProvider gating (hephaestus)", () =>
 
       // #then
       expect(agents.hephaestus).toBeDefined()
-      expect(agents.hephaestus.model).toBe("openai/gpt-5.3-codex")
+      expect(agents.hephaestus.model).toBe("openai/gpt-5.4")
     } finally {
       cacheSpy.mockRestore()
       fetchSpy.mockRestore()

src/shared/model-requirements.ts

@@ -47,11 +47,10 @@ export const AGENT_MODEL_REQUIREMENTS: Record<string, ModelRequirement> = {
   hephaestus: {
     fallbackChain: [
       {
-        providers: ["openai", "venice", "opencode"],
-        model: "gpt-5.3-codex",
+        providers: ["openai", "github-copilot", "venice", "opencode"],
+        model: "gpt-5.4",
         variant: "medium",
       },
-      { providers: ["github-copilot"], model: "gpt-5.4", variant: "medium" },
     ],
     requiresProvider: ["openai", "github-copilot", "venice", "opencode"],
   },