fix(rls): add account_id to AISessionStep creations, fix boards toast

- flowpilot_engine: pass account_id at all 5 AISessionStep instantiation
  sites (_create_step_from_parsed x3, briefing step, status update step).
  Phase 4 RLS blocked every INSERT with NULL account_id — this broke all
  new FlowPilot sessions since the Phase 4 migration was applied.
- integrations: list_boards returns [] on PSAError instead of 502, stopping
  the spurious 'Server error' toast on dashboard load (boards are optional).
- client.ts: 5xx global toast now shows backend detail when available.
- useFlowPilotSession: startSession extracts backend detail for error state;
  suppresses duplicate toast for 5xx (global interceptor already handles it).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

backend/app/api/endpoints/integrations.py

@@ -362,8 +362,9 @@ async def list_boards(
         provider = await get_provider_for_account(current_user.account_id, db)
         boards = await provider.list_boards()
         return [PSABoardResponse(id=b.id, name=b.name) for b in boards]
-    except PSAError as e:
-        raise HTTPException(status_code=502, detail=str(e))
+    except PSAError:
+        # Boards are optional UI chrome — degrade gracefully rather than surfacing a toast
+        return []
 
 
 @router.get("/tickets/search", response_model=list[PSATicketSearchResult])

backend/app/services/flowpilot_engine.py

@@ -330,6 +330,7 @@ async def start_session(
     # 7. Create first step
     step = _create_step_from_parsed(
         session_id=session.id,
+        account_id=session.account_id,
         step_order=0,
         parsed=parsed,
         input_tokens=input_tokens,
@@ -433,6 +434,7 @@ async def process_response(
     # Create new step
     step = _create_step_from_parsed(
         session_id=session.id,
+        account_id=session.account_id,
         step_order=session.step_count - 1,
         parsed=parsed,
         input_tokens=input_tokens,
@@ -694,6 +696,7 @@ async def pickup_session(
     briefing_step = AISessionStep(
         id=uuid.uuid4(),
         session_id=session.id,
+        account_id=session.account_id,
         branch_id=session.active_branch_id if session.is_branching else None,
         step_order=session.step_count,
         step_type="action",
@@ -765,6 +768,7 @@ async def pickup_session(
 
     next_step = _create_step_from_parsed(
         session_id=session.id,
+        account_id=session.account_id,
         step_order=session.step_count - 1,
         parsed=parsed,
         input_tokens=input_tokens,
@@ -997,6 +1001,7 @@ async def generate_status_update(
     step = AISessionStep(
         id=uuid.uuid4(),
         session_id=session.id,
+        account_id=session.account_id,
         branch_id=session.active_branch_id if session.is_branching else None,
         step_order=session.step_count,
         step_type="status_update",
@@ -1440,6 +1445,7 @@ def _format_engineer_response(request: StepResponseRequest) -> str:
 
 def _create_step_from_parsed(
     session_id: UUID,
+    account_id: UUID,
     step_order: int,
     parsed: dict[str, Any],
     input_tokens: int,
@@ -1487,6 +1493,7 @@ def _create_step_from_parsed(
     return AISessionStep(
         id=uuid.uuid4(),
         session_id=session_id,
+        account_id=account_id,
         branch_id=branch_id,
         step_order=step_order,
         step_type=step_type if parsed["type"] != "resolution_suggestion" else "action",

frontend/src/api/client.ts

@@ -49,9 +49,9 @@ function handleGlobalError(error: AxiosError) {
     return
   }
 
-  // Server errors (5xx)
+  // Server errors (5xx) — show backend detail when available, else generic message
   if (status >= 500) {
-    toast.error('Server error - please try again later')
+    toast.error(detail || 'Server error - please try again later')
     return
   }
 }

frontend/src/hooks/useFlowPilotSession.ts

@@ -99,9 +99,14 @@ export function useFlowPilotSession(): UseFlowPilotSession {
       setAllSteps([firstStep])
       setCurrentStep(firstStep)
     } catch (e: unknown) {
-      const message = e instanceof Error ? e.message : 'Failed to start session'
+      // Prefer the backend's detail message over the generic axios status string
+      const detail = (e as any)?.response?.data?.detail
+      const message = typeof detail === 'string' ? detail : (e instanceof Error ? e.message : 'Failed to start session')
       setError(message)
-      toast.error(message)
+      // Global axios interceptor already shows a toast for 5xx — skip duplicate
+      if (!(e as any)?.response?.status || (e as any)?.response?.status < 500) {
+        toast.error(message)
+      }
     } finally {
       setIsLoading(false)
     }