fix: remove irrelevant text-model fields from non-text providers

Remove contextWindow: 0 and maxOutputTokens: 0 from non-text models
(tts, asr, image_gen, video_gen, voice_clone, translation, etc.) and
maxOutputTokens: 0 from embedding/rerank models across kling,
local-whisper, stability, openai, dashscope, google, volcengine,
and zhipu-embedding configs.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

compute/providers/zhipu-embedding.json

@@ -20,7 +20,6 @@
       "serviceType": ["embedding"],
       "description": "智谱嵌入模型v3，支持自定义维度；模型上下文窗口 8K，单条输入最多 3072 tokens。走 OpenAI 兼容 /embeddings 端点（Anthropic 协议不提供 embeddings）",
       "contextWindow": 8192,
-      "maxOutputTokens": 0,
       "capabilities": [
         "text_embedding",
         "semantic_search",
@@ -28,7 +27,8 @@
         "custom_dimensions"
       ],
       "inputPrice": 0.5,
-      "defaultTemperature": 1,
+      "defaultTemperature": null,
+      "defaultTopP": null,
       "extra": {
         "maxInputTokensPerItem": 3072,
         "maxBatchItems": 64,