fix: Refine usage guidelines for actor run tools and clarify handling of widgets (#774)

Author: jirispilka

src/tools/core/call_actor_common.ts

@@ -103,6 +103,7 @@ export function buildCallActorDescription(params: CallActorDescriptionParams): s
             - For a live progress widget the user can watch, call ${HelperTools.ACTOR_CALL_WIDGET} instead.
             - To check status or wait for completion, poll ${HelperTools.ACTOR_RUNS_GET} with the runId.
             - Once the run completes, use ${HelperTools.ACTOR_OUTPUT_GET} tool with the datasetId to fetch full results.
+            - If the Actor name needs resolving first, use ${HelperTools.STORE_SEARCH} (silent). Do NOT use ${HelperTools.STORE_SEARCH_WIDGET} for name resolution.
             - Use dedicated Actor tools when available for better experience
         `);
     } else {

src/tools/core/get_actor_run_common.ts

@@ -27,8 +27,8 @@ The results will include run metadata (status, timestamps), performance stats, a
 USAGE:
 - Use when the user asks about a specific run's status or details.
 - Use to check the status of a run started with call-actor (e.g., before fetching output).
-- If a visual progress widget is available in this session, prefer that tool for UI rendering.
 - Returns pure data with no UI.
+- If \`${HelperTools.ACTOR_CALL_WIDGET}\` or \`${HelperTools.ACTOR_RUNS_GET_WIDGET}\` are available in this session, do NOT call this after them — those render self-polling widgets, additional polling here is forbidden duplicate work.
 
 USAGE EXAMPLES:
 - user_input: Show details of run y2h7sK3Wc (where y2h7sK3Wc is an existing run)