feat: optimize UI-TARS parser with coordinate conversion and action mapping

- Add action mapping for UI-TARS parser to convert action names to option.ActionName
- Implement bounding box to center point coordinate conversion for better accuracy
- Update coordinate normalization to handle coordinates > 1000 properly
- Enhance test cases to verify coordinate scaling and center point conversion
- Improve action argument processing with proper coordinate transformation
- Add comprehensive test coverage for coordinate conversion edge cases

Key improvements:
- Bounding box [x1,y1,x2,y2] now converts to center point [cx,cy] for actions
- Coordinate scaling properly handles different screen resolutions
- Action names are mapped through doubao_1_5_ui_tars_action_mapping
- Enhanced error handling for invalid coordinate formats

uixt/ai/parser_default.go

@@ -21,18 +21,21 @@ func NewLLMContentParser(modelType option.LLMServiceType) LLMContentParser {
 	switch modelType {
 	case option.LLMServiceTypeUITARS:
 		return &UITARSContentParser{
-			systemPrompt: doubao_1_5_ui_tars_planning_prompt,
+			systemPrompt:  doubao_1_5_ui_tars_planning_prompt,
+			actionMapping: doubao_1_5_ui_tars_action_mapping,
 		}
 	default:
 		return &JSONContentParser{
-			systemPrompt: defaultPlanningResponseJsonFormat,
+			systemPrompt:  defaultPlanningResponseJsonFormat,
+			actionMapping: map[string]option.ActionName{},
 		}
 	}
 }
 
 // JSONContentParser parses the response as JSON string format
 type JSONContentParser struct {
-	systemPrompt string
+	systemPrompt  string
+	actionMapping map[string]option.ActionName
 }
 
 func (p *JSONContentParser) SystemPrompt() string {
@@ -83,7 +86,7 @@ func (p *JSONContentParser) Parse(content string, size types.Size) (*PlanningRes
 	}
 
 	// Convert actions to tool calls using function from parser_ui_tars.go
-	toolCalls := convertActionsToToolCalls(normalizedActions)
+	toolCalls := convertActionsToToolCalls(normalizedActions, p.actionMapping)
 
 	return &PlanningResult{
 		ToolCalls:     toolCalls,