feat(dashboard): LLM stats answer who/what spends tokens — member scope chips and spend breakdowns

The stats section was designed single-user: four global charts, no
attribution. Request metadata rows now carry the session's source and
the frame's agent_id/depth (additive ensure_column), denormalized at log
time by the LoggingModel from the owner's pool — off the turn's hot
path, degrading to NULLs, never to a lost row.

The dashboard gains member scope chips filtering every chart, and a
breakdown row splitting the range's billed tokens by member, kind
(chat / sub-agents / cron / system agents / channels), agent, model and
provider — a sub-agent's spend is attributed to the sub-agent itself.
Rows predating the columns group under 'older data'.

web/i18n/en.js

@@ -419,7 +419,6 @@ export default {
   'dashboard.stats.requests': 'Requests {per}',
   'dashboard.stats.tokens':   'Tokens {per}',
   'dashboard.stats.latency':  'Avg latency (ms)',
-  'dashboard.stats.models':   'Models',
   'dashboard.stats.per_min':  '/ min',
   'dashboard.stats.per_hour': '/ hour',
   'dashboard.stats.per_day':  '/ day',
@@ -435,6 +434,26 @@ export default {
   'dashboard.stats.chart.non_cached': 'Non-cached',
   'dashboard.stats.chart.cache_hit':  'Cache hit: {pct}%',
 
+  'dashboard.stats.scope.all':      'Everyone',
+  'dashboard.stats.sub.breakdown':  'How the spend splits in the selected range',
+  'dashboard.stats.by_user':        'By member',
+  'dashboard.stats.by_kind':        'By kind',
+  'dashboard.stats.by_agent':       'By agent',
+  'dashboard.stats.by_model':       'By model',
+  'dashboard.stats.by_provider':    'By provider',
+  'dashboard.stats.user_unknown':   'Unknown',
+  'dashboard.stats.tip.requests':   'requests',
+
+  'dashboard.stats.kind.web':                 'Chat',
+  'dashboard.stats.kind.mobile':              'Mobile',
+  'dashboard.stats.kind.telegram':            'Telegram',
+  'dashboard.stats.kind.cron':                'Scheduled tasks',
+  'dashboard.stats.kind.sub_agent':           'Sub-agents',
+  'dashboard.stats.kind.event_triage':        'Event triage',
+  'dashboard.stats.kind.memory_lint':         'Memory lint',
+  'dashboard.stats.kind.conversation_review': 'Conversation review',
+  'dashboard.stats.kind.unknown':             'Older data',
+
   'dashboard.hero.subtitle':  'Your AI command centre — research, code, plan, and orchestrate. All in one place.',
 
   'dashboard.banner.no_models.title':  'No LLM models configured.',