refactor(ultrawork): remove thinking config injection from model override

Delegate thinking config control to the provider layer rather than
injecting it manually in ultrawork model override.

🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)

src/plugin/ultrawork-model-override.test.ts

@@ -308,10 +308,6 @@ describe("applyUltraworkModelOverrideOnMessage", () => {
     //#then
     expect(output.message.model).toEqual({ providerID: "anthropic", modelID: "claude-opus-4-6" })
     expect(output.message["variant"]).toBe("max")
-    expect(output.message["thinking"]).toEqual({
-      type: "enabled",
-      budgetTokens: 16000,
-    })
     expect(dbOverrideSpy).not.toHaveBeenCalled()
   })
 
@@ -327,10 +323,6 @@ describe("applyUltraworkModelOverrideOnMessage", () => {
     //#then
     expect(output.message.model).toBeUndefined()
     expect(output.message["variant"]).toBe("high")
-    expect(output.message["thinking"]).toEqual({
-      type: "enabled",
-      budgetTokens: 16000,
-    })
     expect(dbOverrideSpy).not.toHaveBeenCalled()
   })