feat: add remix mode for spark table rolling

Introduces a `remix` option that allows each data column in a spark
table to be rolled independently. This is enabled via a `remix=true`
attribute in the directive's extra attributes.

Also refactors spark table scanning by moving markdown parsing logic
from the frontend to a new CLI-side `spark-scanner.ts` utility,
improving separation of concerns.

src/components/journal/completions.ts

@@ -54,6 +54,7 @@ export interface SparkTableCompletion {
   filePath: string;
   csvPath: string;
   headers: string[];
+  remix: boolean;
 }
 
 export interface JournalCompletions {