fix: enable thinking for zhipuai-coding-plan & prevent Korean IME truncation (#22041)

Co-authored-by: claudianus <claudianus@users.noreply.github.com>
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

packages/opencode/test/provider/transform.test.ts

@@ -104,6 +104,58 @@ describe("ProviderTransform.options - setCacheKey", () => {
   })
 })
 
+describe("ProviderTransform.options - zai/zhipuai thinking", () => {
+  const sessionID = "test-session-123"
+
+  const createModel = (providerID: string) =>
+    ({
+      id: `${providerID}/glm-4.6`,
+      providerID,
+      api: {
+        id: "glm-4.6",
+        url: "https://open.bigmodel.cn/api/paas/v4",
+        npm: "@ai-sdk/openai-compatible",
+      },
+      name: "GLM 4.6",
+      capabilities: {
+        temperature: true,
+        reasoning: true,
+        attachment: true,
+        toolcall: true,
+        input: { text: true, audio: false, image: true, video: false, pdf: true },
+        output: { text: true, audio: false, image: false, video: false, pdf: false },
+        interleaved: false,
+      },
+      cost: {
+        input: 0.001,
+        output: 0.002,
+        cache: { read: 0.0001, write: 0.0002 },
+      },
+      limit: {
+        context: 128000,
+        output: 8192,
+      },
+      status: "active",
+      options: {},
+      headers: {},
+    }) as any
+
+  for (const providerID of ["zai-coding-plan", "zai", "zhipuai-coding-plan", "zhipuai"]) {
+    test(`${providerID} should set thinking cfg`, () => {
+      const result = ProviderTransform.options({
+        model: createModel(providerID),
+        sessionID,
+        providerOptions: {},
+      })
+
+      expect(result.thinking).toEqual({
+        type: "enabled",
+        clear_thinking: false,
+      })
+    })
+  }
+})
+
 describe("ProviderTransform.options - google thinkingConfig gating", () => {
   const sessionID = "test-session-123"