fix: migration ModelName, reasoning_content, shell regex, loop boundary

1. migration.go: Set ModelName to userModel when provider matches so
   GetModelConfig(userModel) can find the entry. Previously the migration
   created entries with the provider name as ModelName (e.g. "moonshot")
   but lookup used the model name (e.g. "k2p5"), causing "model not found".

2. openai_compat/provider.go: Preserve reasoning_content in conversation
   history. Thinking models (e.g. Kimi K2, DeepSeek-R1) return
   reasoning_content which must be echoed back. Without it, APIs return
   400: "thinking is enabled but reasoning_content is missing".

3. shell.go: Fix deny pattern regex for format/mkfs/diskpart to use
   (?:^|\s) instead of \b to avoid matching --format flags.
   Fix path extraction regex to use submatch to avoid matching flags
   like -rf as paths.

4. loop.go: Adjust forceCompression mid-point to avoid splitting
   tool-call/result message pairs, which causes API errors.

pkg/config/migration.go

@@ -367,7 +367,9 @@ func ConvertProvidersToModelList(cfg *Config) []ModelConfig {
 
 		// Check if this is the user's configured provider
 		if slices.Contains(m.providerNames, userProvider) && userModel != "" {
-			// Use the user's configured model instead of default
+			// Use the user's configured model instead of default.
+			// Also set ModelName so GetModelConfig(userModel) can find this entry.
+			mc.ModelName = userModel
 			mc.Model = buildModelWithProtocol(m.protocol, userModel)
 		} else if userProvider == "" && userModel != "" && !legacyModelNameApplied {
 			// Legacy config: no explicit provider field but model is specified