Switch to zlib-ng

zlib-ng boasts higher throughput than regular zlib, by combining patches
from Cloudflare, zlib, and ARM's improvements to zlib along with a more
modern codebase.

Profiling consistently shows that compression is the largest CPU expense
by far, so even a minor speed-up here is significant.

native/src/main/c/jni_zlib_common.c

@@ -8,7 +8,7 @@ void JNICALL
 check_zlib_free(JNIEnv *env, z_stream *stream, bool deflate)
 {
     int ret = deflate ? deflateEnd(stream) : inflateEnd(stream);
-    char *msg = stream->msg;
+    const char *msg = stream->msg;
     free((void*) stream);
 
     switch (ret) {

native/src/main/c/jni_zlib_deflate.c

@@ -33,7 +33,7 @@ Java_com_velocitypowered_natives_compression_NativeZlibDeflate_init(JNIEnv *env,
     if (ret == Z_OK) {
         return (jlong) stream;
     } else {
-        char *zlib_msg = stream->msg;
+        const char *zlib_msg = stream->msg;
         free(stream);
         switch (ret) {
             case Z_MEM_ERROR:

native/src/main/c/jni_zlib_inflate.c

@@ -32,7 +32,7 @@ Java_com_velocitypowered_natives_compression_NativeZlibInflate_init(JNIEnv *env,
     if (ret == Z_OK) {
         return (jlong) stream;
     } else {
-        char *zlib_msg = stream->msg;
+        const char *zlib_msg = stream->msg;
         free(stream);
         switch (ret) {
             case Z_MEM_ERROR: