feat: add diagnostic logging for GLM model detection

- Add console logging in getGlmModelOptions() to show when a GLM model is detected
- Log model ID being used in LM Studio and OpenAI-compatible providers
- Log parallel_tool_calls value being applied
- Helps users verify GLM detection is working correctly

Addresses issue #11071 where users cannot verify if GLM detection is functioning

src/api/providers/base-openai-compatible-provider.ts

@@ -79,6 +79,7 @@ export abstract class BaseOpenAiCompatibleProvider<ModelName extends string>
 		// Get model-specific options for GLM models (applies Z.ai optimizations)
 		// This allows third-party GLM models via OpenAI-compatible endpoints to benefit
 		// from the same optimizations used by Z.ai
+		console.log(`[${this.providerName}] Using model ID: "${model}"`)
 		const glmOptions = getGlmModelOptions(model)
 
 		// Centralized cap: clamp to 20% of the context window (unless provider-specific exceptions apply)
@@ -98,6 +99,8 @@ export abstract class BaseOpenAiCompatibleProvider<ModelName extends string>
 			? (metadata?.parallelToolCalls ?? false)
 			: (metadata?.parallelToolCalls ?? true)
 
+		console.log(`[${this.providerName}] parallel_tool_calls set to: ${parallelToolCalls}`)
+
 		// Convert messages with GLM-specific handling when applicable
 		// mergeToolResultText prevents GLM models from dropping reasoning_content
 		const convertedMessages = convertToOpenAiMessages(messages, {

src/api/providers/lm-studio.ts

@@ -45,6 +45,7 @@ export class LmStudioHandler extends BaseProvider implements SingleCompletionHan
 	): ApiStream {
 		// Get model-specific options for GLM models (applies Z.ai optimizations)
 		const modelId = this.getModel().id
+		console.log(`[LM Studio] Using model ID: "${modelId}"`)
 		const glmOptions = getGlmModelOptions(modelId)
 
 		// Convert messages with GLM-specific handling when applicable
@@ -96,6 +97,8 @@ export class LmStudioHandler extends BaseProvider implements SingleCompletionHan
 				? (metadata?.parallelToolCalls ?? false)
 				: (metadata?.parallelToolCalls ?? true)
 
+			console.log(`[LM Studio] parallel_tool_calls set to: ${parallelToolCalls}`)
+
 			const params: OpenAI.Chat.ChatCompletionCreateParamsStreaming & { draft_model?: string } = {
 				model: modelId,
 				messages: openAiMessages,

src/api/providers/utils/model-detection.ts

@@ -72,6 +72,13 @@ export interface GlmModelOptions {
 export function getGlmModelOptions(modelId: string): GlmModelOptions {
 	const isGlm = isGlmModel(modelId)
 
+	// Log GLM model detection result for diagnostics
+	if (isGlm) {
+		console.log(`[GLM Detection] ✓ GLM model detected: "${modelId}"`)
+		console.log(`[GLM Detection]   - mergeToolResultText: true`)
+		console.log(`[GLM Detection]   - disableParallelToolCalls: true`)
+	}
+
 	return {
 		mergeToolResultText: isGlm,
 		disableParallelToolCalls: isGlm,