Fix for leak in thread vector allocation.

In the vvp code generator, when zero/sign extending a sub-expression,
the sub-expression is first evaluated and stored in one location in
the thread vector store, then is copied and extended into a second
location. The storage used for the initial sub-expression evaluation
is never deallocated.

tgt-vvp/eval_expr.c

@@ -1823,6 +1823,8 @@ static struct vector_info draw_pad_expr(ivl_expr_t exp, unsigned wid)
 	    fprintf(vvp_out, "    %%mov %u, 0, %u;\n",
 		    res.base+subv.wid, res.wid - subv.wid);
       }
+      if (subv.base >= 8)
+            clr_vector(subv);
 
       save_expression_lookaside(res.base, exp, wid);
       return res;