Make dlmalloc 2.4x faster for multithreading

This change adds a TLS freelist for small dynamic memory allocations.
Cosmopolitan's TIB is now 512 bytes in size. Single-threaded malloc()
performance isn't impacted by this, until pthread_create() is called.
Single-threaded programs may also want to consider using:

    #include "libc/mem/tinymalloc.inc"

Which will shave 30k off the executable size and sometimes go faster.

libc/errno.h

@@ -29,7 +29,7 @@ COSMOPOLITAN_C_START_
 #define errno                                      \
   (*__extension__({                                \
     errno_t *__ep;                                 \
-    __asm__("sub\t%0,x28,#192-0x3c" : "=r"(__ep)); \
+    __asm__("sub\t%0,x28,#512-0x3c" : "=r"(__ep)); \
     __ep;                                          \
   }))
 #else