feat: add agentic video analysis and clarify action results

Report device actions as dispatched and preserve full failure details. Route
Flash through McpActionExecutor and use ToolFailure across helper tools and
agent loops instead of inferring failure from response text.

Require model-provided descriptions for coordinate targets. Keep observed
element metadata separate from those descriptions in safety checks, visual
summaries, history, and frontend labels.

Add Gemini Interactions API video analysis with configurable processing mode,
chunk size, and timeouts. Centralize Google provider capabilities and token
usage normalization, and update google-genai to 2.22.0.

Preserve system instructions on follow-up interactions, replan pending chunks
when falling back to static processing, and reject unanswered native video
and audio segments instead of caching placeholder results. Keep helper-tool
failure status intact through history tools and committee calls.

Validation: 1,762 Python unit tests passed (3 skipped), 40 frontend formatter
tests passed, and the production UI build completed. Ruff lint, formatting,
the quality ratchet, and git diff checks passed.

apps/showcase_ui/src/app/utils/tool-formatter.util.ts

@@ -911,6 +911,19 @@ export function getToolTitle(tool: any): string {
   }
 }
 
+/**
+ * Join per-point self-described targets (click_sequence `target_descriptions`)
+ * into one label, chained with ' → ' like the coordinate rendering.
+ * Returns '' when the value is missing or holds no usable text.
+ */
+export function joinTargetDescriptions(descriptions: any): string {
+  if (!Array.isArray(descriptions)) return '';
+  const parts = descriptions
+    .map(d => (d === null || d === undefined) ? '' : String(d).trim())
+    .filter(d => d.length > 0);
+  return parts.join(' → ');
+}
+
 /**
  * Get target text description for tools
  */
@@ -950,6 +963,11 @@ export function getToolTargetText(tool: any): string {
   if (name === 'report_failure_analysis') {
     return args.status || args.reason || args.analysis || '';
   }
+  if (name === 'click_sequence' || name === 'tap_sequence') {
+    // One self-described target per point, chained like getToolCoords does.
+    const joined = joinTargetDescriptions(args.target_descriptions);
+    if (joined) return joined;
+  }
   if (args.target && typeof args.target === 'string') {
     return args.target;
   }
@@ -959,7 +977,7 @@ export function getToolTargetText(tool: any): string {
   if (args.index !== undefined) {
     return `Element #${args.index}`;
   }
-  return args.target_text || args.target_class || args.element || args.element_text || (name !== 'input_text' ? args.text : '') || '';
+  return args.target_text || args.target_description || args.target_class || args.element || args.element_text || (name !== 'input_text' ? args.text : '') || '';
 }
 
 /**
@@ -1162,7 +1180,8 @@ export function extractToolExtraParams(toolData: any, cache?: WeakMap<any, Actio
   }
 
   const standardKeys = new Set([
-    'action', 'name', 'type', 'target_text', 'text', 'input_text', 'target',
+    'action', 'name', 'type', 'target_text', 'target_description', 'target_descriptions',
+    'text', 'input_text', 'target',
     'coordinates', 'coords', 'target_bounds', 'bounds', 'target_resource_id',
     'resource_id', 'target_class', 'class_name', 'normalized_coordinates',
     'pre_image_name', 'post_image_name', 'pre_screenshot', 'post_screenshot',