chore(copilot): regenerate the tool catalog for the lean search agent

The search subagent's task description now tells callers to pass a fully
self-contained task — it no longer inherits the conversation (see the
companion mothership change).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Author: j15z

apps/sim/lib/copilot/generated/tool-catalog-v1.ts

@@ -3651,7 +3651,7 @@ export const Search: ToolCatalogEntry = {
     properties: {
       task: {
         description:
-          "One short scoping sentence — the search agent has full conversation context. Example: 'find current Stripe metered-billing API limits' or 'count how many rows in the leads table have invalid emails'.",
+          "A fully self-contained task — the search agent sees none of this conversation, so include the question plus every name, id, constraint, and prior finding it needs. Example: 'find current Stripe metered-billing API limits' or 'count how many rows in the leads table have invalid emails'.",
         type: 'string',
       },
     },

apps/sim/lib/copilot/generated/tool-schemas-v1.ts

@@ -3435,7 +3435,7 @@ export const TOOL_RUNTIME_SCHEMAS: Record<string, ToolRuntimeSchemaEntry> = {
       properties: {
         task: {
           description:
-            "One short scoping sentence — the search agent has full conversation context. Example: 'find current Stripe metered-billing API limits' or 'count how many rows in the leads table have invalid emails'.",
+            "A fully self-contained task — the search agent sees none of this conversation, so include the question plus every name, id, constraint, and prior finding it needs. Example: 'find current Stripe metered-billing API limits' or 'count how many rows in the leads table have invalid emails'.",
           type: 'string',
         },
       },