feat(pipeline): improve new pipeline architecture

- Add TransformDataSource for filtering/mapping source items
- Add MetricsDataSource for rendering live pipeline metrics as ASCII art
- Fix display stage registration in StageRegistry
- Register sources with both class name and simple name aliases
- Fix DisplayStage.init() to pass reuse parameter
- Simplify create_default_pipeline to use DataSourceStage wrapper
- Set pygame as default display
- Remove old pipeline tasks from mise.toml
- Add tests for new pipeline architecture

engine/pipeline/controller.py

@@ -303,18 +303,14 @@ def create_pipeline_from_params(params: PipelineParams) -> Pipeline:
 
 def create_default_pipeline() -> Pipeline:
     """Create a default pipeline with all standard components."""
+    from engine.pipeline.adapters import DataSourceStage
+    from engine.sources_v2 import HeadlinesDataSource
+
     pipeline = Pipeline()
 
-    # Add source stage
-    source = StageRegistry.create("source", "headlines")
-    if source:
-        pipeline.add_stage("source", source)
-
-    # Add effect stages
-    for effect_name in ["noise", "fade", "glitch", "firehose", "hud"]:
-        effect = StageRegistry.create("effect", effect_name)
-        if effect:
-            pipeline.add_stage(f"effect_{effect_name}", effect)
+    # Add source stage (wrapped as Stage)
+    source = HeadlinesDataSource()
+    pipeline.add_stage("source", DataSourceStage(source, name="headlines"))
 
     # Add display stage
     display = StageRegistry.create("display", "terminal")