fix: merge context into fullInput in runEncodeAction to match runGateAction pattern

The context parameter was accepted but never used in runEncodeAction,
silently discarding supplementary context provided by callers. This
restores the fullInput merging pattern used by runGateAction so that
context is included in type detection, artifact parsing, and quality
scoring.

Author: cursoragent

workers/src/orchestrate.ts

@@ -1415,12 +1415,13 @@ async function runEncodeAction(
   state?: OddkitState,
 ): Promise<ActionResult> {
   const startMs = Date.now();
+  const fullInput = context ? `${input}\n${context}` : input;
 
   const types = await discoverEncodingTypes(fetcher, canonUrl);
-  const structured = isStructuredInput(input);
+  const structured = isStructuredInput(fullInput);
   const artifacts = structured
-    ? parseStructuredInput(input, types)
-    : parseUnstructuredInput(input, types);
+    ? parseStructuredInput(fullInput, types)
+    : parseUnstructuredInput(fullInput, types);
 
   // Score each artifact using its type's quality criteria
   const scoredArtifacts = artifacts.map((a) => {