fix: detect models that don't support image input and return clear error (#474)

Some models (Kimi K2, DeepSeek, Qwen text models) don't support image/vision
input. The AI SDK silently drops unsupported image parts, causing confusing
responses where the model acts as if no image was uploaded.

Added supportsImageInput() function to detect unsupported models by name,
and return a 400 error with clear guidance when users try to upload images
to these models.

Closes #469

lib/ai-providers.ts

@@ -906,3 +906,34 @@ export function supportsPromptCaching(modelId: string): boolean {
         modelId.startsWith("eu.anthropic")
     )
 }
+
+/**
+ * Check if a model supports image/vision input.
+ * Some models silently drop image parts without error (AI SDK warning only).
+ */
+export function supportsImageInput(modelId: string): boolean {
+    const lowerModelId = modelId.toLowerCase()
+
+    // Helper to check if model has vision capability indicator
+    const hasVisionIndicator =
+        lowerModelId.includes("vision") || lowerModelId.includes("vl")
+
+    // Models that DON'T support image/vision input (unless vision variant)
+    // Kimi K2 models don't support images
+    if (lowerModelId.includes("kimi") && !hasVisionIndicator) {
+        return false
+    }
+
+    // DeepSeek text models (not vision variants)
+    if (lowerModelId.includes("deepseek") && !hasVisionIndicator) {
+        return false
+    }
+
+    // Qwen text models (not vision variants like qwen-vl)
+    if (lowerModelId.includes("qwen") && !hasVisionIndicator) {
+        return false
+    }
+
+    // Default: assume model supports images
+    return true
+}