SHAKE: in case update is called right after squeezing, permute

Calling update after squeezing is undocumented and non standard,
but if an application still decides to do it, permute the state
before absorbing so that it's still safe to do so.

We can easily do it since we keep track of the state.

Still return an error as this is not the expected usage of SHAKE,
and zeroing the state is another thing we could do.

src/libsodium/crypto_xof/shake128/ref/shake128_ref.c

@@ -28,10 +28,13 @@ shake128_ref_update(shake128_state_internal *state, const unsigned char *in, siz
 {
     size_t consumed = 0;
     size_t chunk_size;
+    int    ret = 0;
 
     if (state->phase != SHAKE128_PHASE_ABSORBING) {
+        crypto_core_keccak1600_permute_24(&state->state);
         state->phase  = SHAKE128_PHASE_ABSORBING;
         state->offset = 0;
+        ret = -1;
     }
 
     while (consumed < inlen) {
@@ -48,7 +51,7 @@ shake128_ref_update(shake128_state_internal *state, const unsigned char *in, siz
         consumed += chunk_size;
     }
 
-    return 0;
+    return ret;
 }
 
 static void

src/libsodium/crypto_xof/shake256/ref/shake256_ref.c

@@ -28,10 +28,13 @@ shake256_ref_update(shake256_state_internal *state, const unsigned char *in, siz
 {
     size_t consumed = 0;
     size_t chunk_size;
+    int    ret = 0;
 
     if (state->phase != SHAKE256_PHASE_ABSORBING) {
+        crypto_core_keccak1600_permute_24(&state->state);
         state->phase  = SHAKE256_PHASE_ABSORBING;
         state->offset = 0;
+        ret = -1;
     }
 
     while (consumed < inlen) {
@@ -48,7 +51,7 @@ shake256_ref_update(shake256_state_internal *state, const unsigned char *in, siz
         consumed += chunk_size;
     }
 
-    return 0;
+    return ret;
 }
 
 static void

src/libsodium/crypto_xof/turboshake128/ref/turboshake128_ref.c

@@ -28,10 +28,13 @@ turboshake128_ref_update(turboshake128_state_internal *state, const unsigned cha
 {
     size_t consumed = 0;
     size_t chunk_size;
+    int    ret = 0;
 
     if (state->phase != TURBOSHAKE128_PHASE_ABSORBING) {
+        crypto_core_keccak1600_permute_12(&state->state);
         state->phase  = TURBOSHAKE128_PHASE_ABSORBING;
         state->offset = 0;
+        ret = -1;
     }
 
     while (consumed < inlen) {
@@ -48,7 +51,7 @@ turboshake128_ref_update(turboshake128_state_internal *state, const unsigned cha
         consumed += chunk_size;
     }
 
-    return 0;
+    return ret;
 }
 
 static void

src/libsodium/crypto_xof/turboshake256/ref/turboshake256_ref.c

@@ -28,10 +28,13 @@ turboshake256_ref_update(turboshake256_state_internal *state, const unsigned cha
 {
     size_t consumed = 0;
     size_t chunk_size;
+    int    ret = 0;
 
     if (state->phase != TURBOSHAKE256_PHASE_ABSORBING) {
+        crypto_core_keccak1600_permute_12(&state->state);
         state->phase  = TURBOSHAKE256_PHASE_ABSORBING;
         state->offset = 0;
+        ret = -1;
     }
 
     while (consumed < inlen) {
@@ -48,7 +51,7 @@ turboshake256_ref_update(turboshake256_state_internal *state, const unsigned cha
         consumed += chunk_size;
     }
 
-    return 0;
+    return ret;
 }
 
 static void