feat(zai): add Z.AI GLM Coding Plan provider preset (#896)

* feat(zai): add Z.AI GLM Coding Plan provider preset

Add dedicated Z.AI provider support for the GLM Coding Plan, enabling
use of GLM-5.1, GLM-5-Turbo, GLM-4.7, and GLM-4.5-Air models through
the OpenAI-compatible shim with proper thinking mode (reasoning_content),
max_tokens handling, and context window sizing.

* fix(zai): unify GLM max output token limits across casing variants

glm-5/glm-4.7 had conservative 16K max output while GLM-5/GLM-4.7
had 131K. Use consistent Z.AI coding plan limits for all GLM variants.

* fix(zai): restore DashScope GLM limits, enable GLM thinking support

- Restore lowercase glm-5/glm-4.7 to 16_384 max output (DashScope limits)
  while keeping Z.AI coding plan high limits on uppercase GLM-* keys only
- Add GLM model support to modelSupportsThinking() so reasoning_content
  is enabled when using GLM-5.x/GLM-4.7 models on Z.AI

* fix(zai): tighten GLM regexes, fix misleading context window comment

- Use precise regex in thinking.ts: exact GLM model matches only,
  no false positives on glm-50/glm-4, includes glm-4.5-air
- Use uppercase-only match in StartupScreen rawModel fallback so
  DashScope lowercase glm-* models aren't mislabeled as Z.AI
- Clarify context window comment: lowercase glm-5.1/glm-5-turbo/
  glm-4.5-air are Z.AI-specific aliases, not DashScope

* fix(zai): scope GLM detection to Z.AI

* improve readability of max_completion_tokens check

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

src/services/api/openaiShim.ts

@@ -64,6 +64,7 @@ import {
 } from './openaiErrorClassification.js'
 import { sanitizeSchemaForOpenAICompat } from '../../utils/schemaSanitizer.js'
 import { redactSecretValueForDisplay } from '../../utils/providerProfile.js'
+import { isZaiBaseUrl } from '../../utils/zaiProvider.js'
 import {
   normalizeToolArguments,
   hasToolFieldMapping,
@@ -93,7 +94,6 @@ const KIMI_CODE_API_HOST = 'api.kimi.com'
 const DEEPSEEK_API_HOSTS = new Set([
   'api.deepseek.com',
 ])
-
 const COPILOT_HEADERS: Record<string, string> = {
   'User-Agent': 'GitHubCopilotChat/0.26.7',
   'Editor-Version': 'vscode/1.99.3',
@@ -1514,7 +1514,8 @@ class OpenAIShimMessages {
       // thinking block we captured on the inbound response.
       preserveReasoningContent:
         isMoonshotCompatibleBaseUrl(request.baseUrl) ||
-        isDeepSeekBaseUrl(request.baseUrl),
+        isDeepSeekBaseUrl(request.baseUrl) ||
+        isZaiBaseUrl(request.baseUrl),
     })
 
     const body: Record<string, unknown> = {
@@ -1553,8 +1554,19 @@ class OpenAIShimMessages {
 
     const isMoonshot = isMoonshotCompatibleBaseUrl(request.baseUrl)
     const isDeepSeek = isDeepSeekBaseUrl(request.baseUrl)
+    const isZai = isZaiBaseUrl(request.baseUrl)
 
-    if ((isGithub || isMistral || isLocal || isMoonshot || isDeepSeek) && body.max_completion_tokens !== undefined) {
+    if (
+      (
+        isGithub ||
+        isMistral ||
+        isLocal ||
+        isMoonshot ||
+        isDeepSeek ||
+        isZai
+      ) &&
+      body.max_completion_tokens !== undefined
+    ) {
       body.max_tokens = body.max_completion_tokens
       delete body.max_completion_tokens
     }
@@ -1562,10 +1574,10 @@ class OpenAIShimMessages {
     // mistral and gemini don't recognize body.store — Gemini returns 400
     // "Invalid JSON payload received. Unknown name 'store': Cannot find field."
     // Moonshot direct API, Kimi Code's OpenAI-compatible coding endpoint,
-    // and DeepSeek have not published support for the parameter either;
+    // DeepSeek, and Z.AI have not published support for the parameter either;
     // strip it preemptively to avoid the same class of error on strict-parse
     // providers.
-    if (isMistral || isGeminiMode() || isMoonshot || isDeepSeek) {
+    if (isMistral || isGeminiMode() || isMoonshot || isDeepSeek || isZai) {
       delete body.store
     }
 
@@ -1593,6 +1605,17 @@ class OpenAIShimMessages {
       }
     }
 
+    // Z.AI uses the same thinking format as DeepSeek: { type: "enabled" | "disabled" }
+    // with reasoning_content in responses.
+    if (isZai) {
+      const requestedThinkingType = (params.thinking as { type?: string } | undefined)?.type
+      if (requestedThinkingType && requestedThinkingType !== 'disabled') {
+        body.thinking = { type: 'enabled' }
+      } else if (requestedThinkingType === 'disabled') {
+        body.thinking = { type: 'disabled' }
+      }
+    }
+
     if (params.tools && params.tools.length > 0) {
       const converted = convertTools(
         params.tools as Array<{