refactor: centralize CPU core detection using cgroup-aware utility

Introduce `obisys::effective_parallelism()` to read Linux cgroup v1/v2 CPU quotas from sysfs, preventing thread pool oversubscription in containerized environments. Replace direct `std::thread::available_parallelism()` calls across `obikindex` and `obikmer` with this centralized function. Bump `obikmer` version to 1.1.41.

src/obikindex/src/numa.rs

@@ -79,9 +79,7 @@ pub fn build() -> NumaSetup {
     }
 
     // UMA fallback: single synthetic node, all cores, no pool, no pinning.
-    let n_cores = std::thread::available_parallelism()
-        .map(|n| n.get())
-        .unwrap_or(1);
+    let n_cores = obisys::effective_parallelism();
     debug!("UMA: single synthetic node, {} core(s)", n_cores);
     NumaSetup {
         pools: vec![None],
@@ -91,9 +89,7 @@ pub fn build() -> NumaSetup {
 
 #[cfg(not(feature = "numa"))]
 pub fn build() -> NumaSetup {
-    let n_cores = std::thread::available_parallelism()
-        .map(|n| n.get())
-        .unwrap_or(1);
+    let n_cores = obisys::effective_parallelism();
     debug!("UMA: single synthetic node, {} core(s)", n_cores);
     NumaSetup {
         pools: vec![None],

src/obikindex/src/siblings.rs

@@ -350,7 +350,7 @@ impl KmerIndex {
         // every single variant — fine at the scale of a handful of test
         // k-mers, but with billions of lookups against a real index this
         // manifested as ~90% system time, observed in practice. ───────────
-        let n_workers = std::thread::available_parallelism().map(|n| n.get()).unwrap_or(4);
+        let n_workers = obisys::effective_parallelism();
         let capacity = 256;
 
         // Throttling is not optional once a `Flat` stage is in the pipeline