fix(util): do not strip thinking suffix on registered models

NormalizeThinkingModel now checks ModelSupportsThinking before removing
"-thinking" or "-thinking-<ver>", avoiding accidental parsing of model
names where the suffix is part of the official id (e.g., kimi-k2-thinking,
qwen3-235b-a22b-thinking-2507).

The registry adds ThinkingSupport metadata for several models and
propagates it via ModelInfo (e.g., kimi-k2-thinking, deepseek-r1,
qwen3-235b-a22b-thinking-2507, minimax-m2), enabling accurate detection
of thinking-capable models and correcting base model inference.

internal/registry/model_definitions.go

@@ -623,6 +623,7 @@ func GetIFlowModels() []*ModelInfo {
 		DisplayName string
 		Description string
 		Created     int64
+		Thinking    *ThinkingSupport
 	}{
 		{ID: "tstars2.0", DisplayName: "TStars-2.0", Description: "iFlow TStars-2.0 multimodal assistant", Created: 1746489600},
 		{ID: "qwen3-coder-plus", DisplayName: "Qwen3-Coder-Plus", Description: "Qwen3 Coder Plus code generation", Created: 1753228800},
@@ -632,17 +633,17 @@ func GetIFlowModels() []*ModelInfo {
 		{ID: "kimi-k2-0905", DisplayName: "Kimi-K2-Instruct-0905", Description: "Moonshot Kimi K2 instruct 0905", Created: 1757030400},
 		{ID: "glm-4.6", DisplayName: "GLM-4.6", Description: "Zhipu GLM 4.6 general model", Created: 1759190400},
 		{ID: "kimi-k2", DisplayName: "Kimi-K2", Description: "Moonshot Kimi K2 general model", Created: 1752192000},
-		{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 general model", Created: 1762387200},
+		{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 thinking model", Created: 1762387200, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
 		{ID: "deepseek-v3.2-chat", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2", Created: 1764576000},
 		{ID: "deepseek-v3.2", DisplayName: "DeepSeek-V3.2-Exp", Description: "DeepSeek V3.2 experimental", Created: 1759104000},
 		{ID: "deepseek-v3.1", DisplayName: "DeepSeek-V3.1-Terminus", Description: "DeepSeek V3.1 Terminus", Created: 1756339200},
-		{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200},
+		{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
 		{ID: "deepseek-v3", DisplayName: "DeepSeek-V3-671B", Description: "DeepSeek V3 671B", Created: 1734307200},
 		{ID: "qwen3-32b", DisplayName: "Qwen3-32B", Description: "Qwen3 32B", Created: 1747094400},
-		{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600},
+		{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
 		{ID: "qwen3-235b-a22b-instruct", DisplayName: "Qwen3-235B-A22B-Instruct", Description: "Qwen3 235B A22B Instruct", Created: 1753401600},
 		{ID: "qwen3-235b", DisplayName: "Qwen3-235B-A22B", Description: "Qwen3 235B A22B", Created: 1753401600},
-		{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000},
+		{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
 	}
 	models := make([]*ModelInfo, 0, len(entries))
 	for _, entry := range entries {
@@ -654,6 +655,7 @@ func GetIFlowModels() []*ModelInfo {
 			Type:        "iflow",
 			DisplayName: entry.DisplayName,
 			Description: entry.Description,
+			Thinking:    entry.Thinking,
 		})
 	}
 	return models

internal/util/thinking_suffix.go

@@ -52,6 +52,11 @@ func NormalizeThinkingModel(modelName string) (string, map[string]any) {
 		matched = true
 	default:
 		if idx := strings.LastIndex(lower, "-thinking-"); idx != -1 {
+			// Skip stripping if the original model is a registered thinking model.
+			// This prevents "-thinking-2507" in "qwen3-235b-a22b-thinking-2507" from being parsed.
+			if ModelSupportsThinking(modelName) {
+				break
+			}
 			value := modelName[idx+len("-thinking-"):]
 			if value != "" {
 				if parsed, ok := parseIntPrefix(value); ok {
@@ -95,10 +100,16 @@ func NormalizeThinkingModel(modelName string) (string, map[string]any) {
 				}
 			}
 		} else if strings.HasSuffix(lower, "-thinking") {
-			baseModel = modelName[:len(modelName)-len("-thinking")]
+			candidateBase := modelName[:len(modelName)-len("-thinking")]
-			effort := "medium"
+			// Only strip the suffix if the original model is NOT a registered thinking model.
-			reasoningEffort = &effort
+			// This prevents stripping "-thinking" from models like "kimi-k2-thinking" where
-			matched = true
+			// the suffix is part of the model's actual name.
+			if !ModelSupportsThinking(modelName) {
+				baseModel = candidateBase
+				effort := "medium"
+				reasoningEffort = &effort
+				matched = true
+			}
 		}
 	}