tunnel_to: shrink huge buffer size

the buffer buff was only used for the initial handshake packets,
which in all supported protocols are usual less than a 100 bytes,
with user/pass and dns name at maximum we'd require 768 bytes,
which still leaves us a formidable 256 bytes for the rest of
the packet.
this fixes a segfault with microsocks which on musl uses tiny
thread stack sizes of 8KB.

src/core.c

@@ -188,6 +188,7 @@ static int timed_connect(int sock, const struct sockaddr *addr, socklen_t len) {
 
 
 #define INVALID_INDEX 0xFFFFFFFFU
+#define BUFF_SIZE 1024  // used to read responses from proxies.
 static int tunnel_to(int sock, ip_type ip, unsigned short port, proxy_type pt, char *user, char *pass) {
 	char *dns_name = NULL;
 	char hostnamebuf[MSG_LEN_MAX];

src/core.h

@@ -23,7 +23,6 @@
 
 #ifndef __CORE_HEADER
 #define __CORE_HEADER
-#define BUFF_SIZE 8*1024  // used to read responses from proxies.
 #define     MAX_LOCALNET 64
 #define     MAX_DNAT 64