Fix the build I broke, which CI caught first

`Share.carried` became `Share.carrying` when the per-pass count was separated
from the total, and the JNI still read the old name. `diaswarm-android` did not
compile.

I renamed the field, re-ran `diaswarm-net`, and pushed — never re-running the
crate that consumes it. The crates workflow added this morning failed in 2m13s
on exactly that, which is the first time it has caught anything and the reason
it exists: before today `diaswarm-net` and `diaswarm-android` had never run in
CI at all, and this would have sat broken on main until somebody built an APK.

All four crates run locally this time: core 58, keys 27, android 17, net 45,
0 failed.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: alex-poor

crates/diaswarm-android/src/lib.rs

@@ -2486,7 +2486,7 @@ pub extern "system" fn Java_nz_diaswarm_jni_SwarmNative_keysCarryAll<'a>(
         &own,
         max_adopt.max(0) as usize,
     )) {
-        Ok(share) => share.carried as jlong,
+        Ok(share) => share.carrying as jlong,
         Err(_) => -4,
     }
 }