feat: default two structured re-asks and sharper AIMD cut

Round 5 left three residual failure classes; ladder exhaustion (3.2%
of structured calls with a single re-ask) and 429 leakage (5.8%, AIMD
oscillating above the sustainable point) are addressable: re-ask
default goes 1 to 2 (conservative vs instructor's 3) and the AIMD cut
factor drops to 0.5.

.env.example

@@ -54,7 +54,7 @@ PGW_TELEMETRY_BACKEND=none               # sqlite | postgres | none(必填)
 # PGW_PRICING_PATH=config/prices.json           # 可选: {"<model>": {"input_per_1m": x, "output_per_1m": y}};缺省 cost 恒 None
 # PGW_CACHE_NAMESPACE=<项目名或租户前缀>          # 缓存启用时必填(防跨项目毒化)
 # PGW_CACHE_TTL_S=604800                        # 缓存启用时必填,须 > 0
-# PGW_STRUCTURED_MAX_RETRIES=1                  # 0 = 解析失败不重问(CHS 策略)
+# PGW_STRUCTURED_MAX_RETRIES=2                  # 缺省 2(M2.5);0 = 解析失败不重问(CHS 策略)
 # PGW_LEASE_TTL_S=1500                          # permit 租约;须 ≥ 最大源 timeout
 
 # ══ Redis(缓存 + 分布式限流/熔断)══