feat: progressive drill-down dashboard with RED metrics and SLA compliance (#94)

Three-level dashboard driven by sidebar selection:
- L1 (no selection): all-apps overview with health table, per-app charts
- L2 (app selected): route performance table, error velocity, top errors
- L3 (route selected): processor table, latency heatmap data, bottleneck KPI

Backend: 3 new endpoints (timeseries/by-app, timeseries/by-route, errors/top),
per-app SLA settings (app_settings table, V12 migration), exact SLA compliance
from executions hypertable, error velocity with acceleration detection.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

cameleer3-server-app/src/main/resources/db/migration/V12__app_settings.sql

@@ -0,0 +1,11 @@
+-- Per-application dashboard settings (SLA thresholds, health dot thresholds)
+CREATE TABLE app_settings (
+    app_id            TEXT             PRIMARY KEY,
+    sla_threshold_ms  INTEGER          NOT NULL DEFAULT 300,
+    health_error_warn DOUBLE PRECISION NOT NULL DEFAULT 1.0,
+    health_error_crit DOUBLE PRECISION NOT NULL DEFAULT 5.0,
+    health_sla_warn   DOUBLE PRECISION NOT NULL DEFAULT 99.0,
+    health_sla_crit   DOUBLE PRECISION NOT NULL DEFAULT 95.0,
+    created_at        TIMESTAMPTZ      NOT NULL DEFAULT now(),
+    updated_at        TIMESTAMPTZ      NOT NULL DEFAULT now()
+);