refactor(meta-service0): rotbl: use spawn_blocking() instead blocking_in_place()

`blocking_in_place()` may still exhaust the worker threads.

Author: drmingdrmer

Cargo.lock

@@ -460,7 +460,7 @@ reqwest-hickory-resolver = "0.2"
 ringbuffer = "0.14.2"
 rmp-serde = "1.1.1"
 roaring = { version = "^0.10", features = ["serde"] }
-rotbl = { version = "0.2.8", features = [] }
+rotbl = { version = "0.2.9", features = [] }
 rust_decimal = "1.26"
 rustix = "0.38.37"
 rustls = { version = "0.23.27", features = ["ring", "tls12"], default-features = false }

Cargo.toml

@@ -213,6 +213,7 @@ mod tests {
 
         // blocking_in_place sleep
 
+        #[allow(clippy::disallowed_methods)]
         let f = async move {
             tokio::task::spawn_blocking(|| {
                 std::thread::sleep(Duration::from_millis(100));