replace ContainerRuntime CLI to Bollard crate (default: docker socket)

limit tool result to open router tool response (with truncated info for
ai)

Hard kill if graceful shutdown is too long

Cargo.toml

@@ -8,6 +8,7 @@ resolver = "3"
 [workspace.dependencies]
 reqwest = { version = "0.13", default-features = false, features = ["json", "native-tls", "stream"] }
 tokio = { version = "1.53", features = ["full"] }
+tokio-stream = "0.1"
 tokio-util = "0.7"
 futures-util = "0.3"
 serde_json = "1.0"
@@ -24,6 +25,7 @@ anyhow = { version = "1", features = ["backtrace"] }
 thiserror = "2.0"
 ring = "0.17"
 hex = "0.4"
+bytes = "1.1"
 metrics = "0.24"
 metrics-exporter-prometheus = { version = "0.18", default-features = false, features = ["http-listener"] }