docs: align jt808 mileage storage

docs/target-architecture.md

@@ -126,10 +126,11 @@ calculation_method = JT808_TOTAL_MILEAGE_DIFF
 
 - Store the metric in `vehicle_stat_metric` with
   `metric_key = daily_mileage_km`.
-- Keep rolling daily calculation state in Redis.
 
 There is no JT808-specific daily mileage table and no in-memory production
-state-store mode.
+state-store mode. Daily start total, latest total, and derived mileage are all
+stored in `vehicle_stat_metric`, so restart recovery does not require a separate
+Redis mileage state.
 
 ### Latest State
 
@@ -156,7 +157,6 @@ flowchart LR
   kafka --> state["vehicle-state-service"]
   history --> tdengine["TDengine<br/>raw_frames + locations"]
   analytics --> mysql["MySQL<br/>vehicle_stat_metric"]
-  analytics --> redisCalc["Redis<br/>mileage state"]
   state --> redisState["Redis<br/>latest state"]
 ```
 
@@ -167,7 +167,8 @@ flowchart LR
 - Kafka producer failures are handled by sink retry/circuit-breaker behavior and
   DLQ topics where configured.
 - History consumers should commit offsets only after TDengine writes succeed.
-- Analytics state can recover from Redis state or Kafka replay.
+- Analytics daily mileage state can recover from `vehicle_stat_metric` or Kafka
+  replay.
 - Raw archive failures must be observable, but event publication and history
   indexing remain separate concerns.
 
@@ -182,6 +183,6 @@ flowchart LR
 - Location queries page over compact rows and reference raw records instead of
   duplicating full raw JSON.
 - JT808 daily mileage is stored only in `vehicle_stat_metric`.
-- Runtime state needed for 808 mileage uses Redis, with no production memory
-  mode.
+- Runtime state needed for 808 mileage is also in `vehicle_stat_metric`, with no
+  Redis or production memory state-store mode.
 - Build validation covers the active modules and their composition tests.