perf(rust): make mimalloc opt-in behind a cargo feature

pydantic-core shipped mimalloc from Jun 2022 to Aug 2023 and removed it
(pydantic-core#900) after: a TLS-linkage fix (pydantic-core#363), manylinux
cross-compile breakage (pydantic-core#893), 1GiB of reserved virtual memory
at import breaking RLIMIT_AS-limited processes (pydantic#7167), and
container memory-footprint complaints (pydantic#6620). litellm wheels land
in exactly those memory-limited container deployments, so the allocator
override must be an explicit build-time choice, not a default imposed on
every wheel. The feature stays available for self-built deployments that
want it; wheels build without it.

litellm-rust/crates/python-bridge/Cargo.toml

@@ -13,11 +13,12 @@ crate-type = ["cdylib"]
 default = ["abi3"]
 abi3 = ["pyo3/abi3-py310"]
 extension-module = ["pyo3/extension-module"]
+mimalloc = ["dep:mimalloc"]
 panic-test = []
 
 [dependencies]
 futures-util.workspace = true
-mimalloc.workspace = true
+mimalloc = { workspace = true, optional = true }
 tracing.workspace = true
 tracing-subscriber.workspace = true
 litellm-core = { workspace = true, features = ["bedrock-auth"] }

litellm-rust/crates/python-bridge/src/lib.rs

@@ -11,6 +11,7 @@ use pyo3::prelude::*;
 use pyo3::types::PyAny;
 use serde_json::Value;
 
+#[cfg(feature = "mimalloc")]
 #[global_allocator]
 static ALLOCATOR: mimalloc::MiMalloc = mimalloc::MiMalloc;