docs: update load balancing strategies to include optimal-latency option in API and configuration documentation

Author: 𝐘𝐨𝐬𝐞𝐛𝐲𝐭𝐞

docs/en/api.md

@@ -1588,7 +1588,7 @@ Examples:
 | `key` | Private key path | File path | None | Server only |
 | `dns` | DNS cache duration | Time duration (e.g., `5m`, `30s`, `1h`) | `5m` | Both |
 | `sni` | Server Name Indication | Hostname | `none` | Client dual-end handshake mode only |
-| `lbs` | Load balancing strategy | `0`(round-robin), `1`(sticky-failover), `2`(primary-backup) | `0` | Both | 
+| `lbs` | Load balancing strategy | `0`(round-robin), `1`(optimal-latency), `2`(primary-backup) | `0` | Both | 
 | `min` | Minimum pool capacity | Integer > 0 | `64` | Client dual-end handshake mode only |
 | `max` | Maximum pool capacity | Integer > 0 | `1024` | Dual-end handshake mode |
 | `mode` | Runtime mode control | `0`(auto), `1`(force mode 1), `2`(force mode 2) | `0` | Both |

docs/en/configuration.md

@@ -699,10 +699,11 @@ NodePass provides three load balancing strategies controlled by the `lbs` parame
 - **Failover**: When a connection to an address fails, immediately tries the next address to ensure service availability
 - **Automatic Recovery**: Failed addresses are retried in subsequent rotation cycles and automatically resume receiving traffic after recovery
 
-**Strategy 1 (Sticky-Failover):**
-- **Connection Stickiness**: Continues using the current target address as long as connections succeed
-- **Failover Only**: Switches to the next address only when connection failures occur
-- **Persistent Target**: Once switched to a new address, all subsequent connections use that address until it fails
+**Strategy 1 (Optimal-Latency):**
+- **Intelligent Routing**: Periodically probes targets and automatically selects the one with the lowest latency for connections.
+- **Sticky Selection**: Once the optimal target is chosen, subsequent connections within the cycle preferentially use that target.
+- **Automatic Filtering**: Unhealthy targets are automatically excluded from routing and re-evaluated only after recovery.
+- **Failover**: If the optimal target fails, other targets are tried in order to ensure successful connections.
 
 **Strategy 2 (Primary-Backup):**
 - **Priority-Based**: Always attempts to connect to the first address (primary); only uses backups when primary fails
@@ -716,7 +717,7 @@ Example configurations:
 # Round-robin (lbs=0, cycles through targets on each connection)
 nodepass "server://0.0.0.0:10101/backend1:8080,backend2:8080,backend3:8080?lbs=0"
 
-# Sticky failover (lbs=1, stays on current target until failure)
+# Optimal-latency (lbs=1, automatically routes to fastest target)
 nodepass "server://0.0.0.0:10101/backend1:8080,backend2:8080,backend3:8080?lbs=1"
 
 # Primary-backup (lbs=2, primary priority and scheduled fallback)
@@ -729,7 +730,7 @@ nodepass "server://0.0.0.0:10101/main.com:443,spare1.com:443,spare2.com:443?lbs=
 
 Choose the appropriate strategy based on your needs:
 - **Use lbs=0** for even load distribution across all backends
-- **Use lbs=1** for session persistence and reduced backend switching
+- **Use lbs=1** for intelligent routing to the lowest latency target
 - **Use lbs=2** for primary-backup scenarios with automatic failback
 
 ### Use Cases

docs/zh/api.md

@@ -1588,7 +1588,7 @@ client://<server_host>:<server_port>/<local_host>:<local_port>?<parameters>
 | `key` | 私钥路径 | 文件路径 | 无 | 仅服务端 |
 | `dns` | DNS缓存时间 | 时间长度 (如 `10m`, `30s`, `1h`) | `5m` | 两者 |
 | `sni` | 主机名指示 | 主机名 | `none` | 仅客户端双端握手模式 |
-| `lbs` | 负载均衡策略 | `0`(轮询), `1`(粘性), `2`(主备) | `0` | 两者 |
+| `lbs` | 负载均衡策略 | `0`(轮询转移), `1`(最优延迟), `2`(主备回落) | `0` | 两者 |
 | `min` | 最小连接池容量 | 整数 > 0 | `64` | 仅客户端双端握手模式 |
 | `max` | 最大连接池容量 | 整数 > 0 | `1024` | 双端握手模式 |
 | `mode` | 运行模式控制 | `0`(自动), `1`(强制模式1), `2`(强制模式2) | `0` | 两者 |

docs/zh/configuration.md

@@ -694,17 +694,18 @@ nodepass "client://127.0.0.1:1080/app1.local:8080,app2.local:8080?mode=1"
 
 NodePass提供三种负载均衡策略，通过 `lbs` 参数控制：
 
-**策略0（轮询故障转移）：**
+**策略0（轮询转移）：**
 - **负载均衡**：每次成功建立连接后，自动切换到下一个目标地址，实现流量均匀分布
 - **故障转移**：当某个地址连接失败时，立即尝试下一个地址，确保服务高可用
 - **自动恢复**：失败的地址会在轮询周期中重新尝试，故障恢复后自动接入流量
 
-**策略1（粘性故障转移）：**
-- **连接粘性**：只要连接成功，就持续使用当前目标地址
-- **失败切换**：只有当连接失败时才切换到下一个地址
-- **目标持久**：切换到新地址后，后续所有连接都使用该地址直到失败
+**策略1（最优延迟）：**
+- **智能路由**：周期性探测，自动选择延迟最低的目标进行连接
+- **粘性选择**：最优目标选择后，周期内后续连接优先使用该目标
+- **自动筛选**：失效目标自动排除，不参与路由，直至恢复后再评估
+- **故障转移**：最优目标失败时按顺序尝试其他目标，确保连接成功
 
-**策略2（主备故障转移）：**
+**策略2（主备回落）：**
 - **主备优先**：首个目标地址主用，后续目标地址备用，按顺序优先级递减
 - **故障转移**：主目标地址失败时按顺序尝试备用地址，成功后保持粘性
 - **定时回落**：每隔固定时间自动尝试回落到主用目标地址
@@ -713,13 +714,13 @@ NodePass提供三种负载均衡策略，通过 `lbs` 参数控制：
 配置示例：
 
 ```bash
-# 轮询故障转移（lbs=0，每次连接切换目标）
+# 轮询转移（lbs=0，每次连接切换目标）
 nodepass "server://0.0.0.0:10101/backend1:8080,backend2:8080,backend3:8080?lbs=0"
 
-# 粘性故障转移（lbs=1，成功时保持当前目标）
+# 最优延迟（lbs=1，自动路由到最快目标）
 nodepass "server://0.0.0.0:10101/backend1:8080,backend2:8080,backend3:8080?lbs=1"
 
-# 主备故障转移（lbs=2，首目标优先且定时回落）
+# 主备回落（lbs=2，首目标优先且定时回落）
 nodepass "server://0.0.0.0:10101/primary:8080,backup1:8080,backup2:8080?lbs=2"
 
 # 自定义回落间隔为2分钟
@@ -729,7 +730,7 @@ nodepass "server://0.0.0.0:10101/main.com:443,spare1.com:443,spare2.com:443?lbs=
 
 根据实际需求选择合适的策略：
 - **使用 lbs=0** 实现所有后端的均匀负载分布
-- **使用 lbs=1** 减少后端切换，保持会话粘性
+- **使用 lbs=1** 智能路由到延迟最低目标地址
 - **使用 lbs=2** 主备模式，故障消除自动回切
 
 ### 使用场景