various: Fix -fsanitize=undefined errors.

[messages truncated to pass ci_commit_formatting_run]
[the line numbers match 0e87459]
```
../../extmod/crypto-algorithms/sha256.c:49:19: runtime error: left shif...
../../extmod/moduasyncio.c:106:35: runtime error: member access within ...
../../py/binary.c:210:13: runtime error: left shift of negative value -...
../../py/mpz.c:744:16: runtime error: negation of -9223372036854775808 ...
../../py/objint.c:109:22: runtime error: left shift of 1 by 31 places c...
../../py/objint_mpz.c:374:9: runtime error: left shift of 4611686018427...
../../py/objint_mpz.c:374:9: runtime error: left shift of negative valu...
../../py/parsenum.c:106:14: runtime error: left shift of 46116860184273...
../../py/runtime.c:395:33: runtime error: left shift of negative value ...
../../py/showbc.c:177:28: runtime error: left shift of negative value -...
../../py/vm.c:321:36: runtime error: left shift of negative value -1```

My testing was done on an amd64 debian buster system using gcc-8.3 and
these settings:
```
CFLAGS += -g3 -Og -fsanitize=undefined
LDFLAGS += -fsanitize=undefined
```

The introduced PAIRHEAP macro's conditional (x ? &x->i : NULL)
assembles (under amd64 gcc 8.3 -Os) to the same as &x->i, since i is
the initial field of the struct. However, for the purposes of undefined
behavior analysis the conditional is needed.

Signed-off-by: Jeff Epler <jepler@gmail.com>

extmod/crypto-algorithms/sha256.c

@@ -46,7 +46,7 @@ static void sha256_transform(CRYAL_SHA256_CTX *ctx, const BYTE data[])
 	WORD a, b, c, d, e, f, g, h, i, j, t1, t2, m[64];
 
 	for (i = 0, j = 0; i < 16; ++i, j += 4)
-		m[i] = (data[j] << 24) | (data[j + 1] << 16) | (data[j + 2] << 8) | (data[j + 3]);
+		m[i] = ((uint32_t)data[j] << 24) | (data[j + 1] << 16) | (data[j + 2] << 8) | (data[j + 3]);
 	for ( ; i < 64; ++i)
 		m[i] = SIG1(m[i - 2]) + m[i - 7] + SIG0(m[i - 15]) + m[i - 16];
 

extmod/moduasyncio.c

@@ -40,6 +40,8 @@ typedef struct _mp_obj_task_t {
     mp_obj_t ph_key;
 } mp_obj_task_t;
 
+#define PAIRHEAP(x) ((x) ? &x->pairheap : NULL)
+
 typedef struct _mp_obj_task_queue_t {
     mp_obj_base_t base;
     mp_obj_task_t *heap;
@@ -103,7 +105,7 @@ STATIC mp_obj_t task_queue_push_sorted(size_t n_args, const mp_obj_t *args) {
         assert(mp_obj_is_small_int(args[2]));
         task->ph_key = args[2];
     }
-    self->heap = (mp_obj_task_t *)mp_pairheap_push(task_lt, &self->heap->pairheap, &task->pairheap);
+    self->heap = (mp_obj_task_t *)mp_pairheap_push(task_lt, PAIRHEAP(self->heap), PAIRHEAP(task));
     return mp_const_none;
 }
 STATIC MP_DEFINE_CONST_FUN_OBJ_VAR_BETWEEN(task_queue_push_sorted_obj, 2, 3, task_queue_push_sorted);

py/binary.c

@@ -202,7 +202,7 @@ long long mp_binary_get_int(size_t size, bool is_signed, bool big_endian, const
         delta = 1;
     }
 
-    long long val = 0;
+    unsigned long long val = 0;
     if (is_signed && *src & 0x80) {
         val = -1;
     }

py/mpz.c

@@ -741,7 +741,7 @@ void mpz_set_from_ll(mpz_t *z, long long val, bool is_signed) {
     unsigned long long uval;
     if (is_signed && val < 0) {
         z->neg = 1;
-        uval = -val;
+        uval = -(unsigned long long)val;
     } else {
         z->neg = 0;
         uval = val;

py/objint.c

@@ -106,14 +106,14 @@ STATIC mp_fp_as_int_class_t mp_classify_fp_as_int(mp_float_t val) {
         #if MICROPY_FLOAT_IMPL == MICROPY_FLOAT_IMPL_DOUBLE
         e |= u.i[MP_ENDIANNESS_BIG] != 0;
         #endif
-        if ((e & ~(1 << MP_FLOAT_SIGN_SHIFT_I32)) == 0) {
+        if ((e & ~(1U << MP_FLOAT_SIGN_SHIFT_I32)) == 0) {
             // handle case of -0 (when sign is set but rest of bits are zero)
             e = 0;
         } else {
-            e += ((1 << MP_FLOAT_EXP_BITS) - 1) << MP_FLOAT_EXP_SHIFT_I32;
+            e += ((1U << MP_FLOAT_EXP_BITS) - 1) << MP_FLOAT_EXP_SHIFT_I32;
         }
     } else {
-        e &= ~((1 << MP_FLOAT_EXP_SHIFT_I32) - 1);
+        e &= ~((1U << MP_FLOAT_EXP_SHIFT_I32) - 1);
     }
     // 8 * sizeof(uintptr_t) counts the number of bits for a small int
     // TODO provide a way to configure this properly

py/runtime.c

@@ -392,7 +392,7 @@ mp_obj_t mp_binary_op(mp_binary_op_t op, mp_obj_t lhs, mp_obj_t rhs) {
                         goto generic_binary_op;
                     } else {
                         // use standard precision
-                        lhs_val <<= rhs_val;
+                        lhs_val = (mp_uint_t)lhs_val << rhs_val;
                     }
                     break;
                 }

py/showbc.c

@@ -174,7 +174,7 @@ const byte *mp_bytecode_print_str(const mp_print_t *print, const byte *ip) {
                 num--;
             }
             do {
-                num = (num << 7) | (*ip & 0x7f);
+                num = ((mp_uint_t)num << 7) | (*ip & 0x7f);
             } while ((*ip++ & 0x80) != 0);
             mp_printf(print, "LOAD_CONST_SMALL_INT " INT_FMT, num);
             break;

py/smallint.h

@@ -37,7 +37,7 @@
 #if MICROPY_OBJ_REPR == MICROPY_OBJ_REPR_A || MICROPY_OBJ_REPR == MICROPY_OBJ_REPR_C
 
 #define MP_SMALL_INT_MIN ((mp_int_t)(((mp_int_t)MP_OBJ_WORD_MSBIT_HIGH) >> 1))
-#define MP_SMALL_INT_FITS(n) ((((n) ^ ((n) << 1)) & MP_OBJ_WORD_MSBIT_HIGH) == 0)
+#define MP_SMALL_INT_FITS(n) ((((n) ^ ((mp_uint_t)(n) << 1)) & MP_OBJ_WORD_MSBIT_HIGH) == 0)
 // Mask to truncate mp_int_t to positive value
 #define MP_SMALL_INT_POSITIVE_MASK ~(MP_OBJ_WORD_MSBIT_HIGH | (MP_OBJ_WORD_MSBIT_HIGH >> 1))
 

py/vm.c

@@ -312,7 +312,7 @@ FRAME_SETUP();
                     DISPATCH();
 
                 ENTRY(MP_BC_LOAD_CONST_SMALL_INT): {
-                    mp_int_t num = 0;
+                    mp_uint_t num = 0;
                     if ((ip[0] & 0x40) != 0) {
                         // Number is negative
                         num--;