Fuzz HRP of bech32 as well

Also separated the roundtrip testing from the random string decoding for clarity

Note that while BIP 173 claims:
```
The human-readable part, which is intended to convey the type of data, or anything else that is relevant to the reader. This part MUST contain 1 to 83 US-ASCII characters, with each character having a value in the range [33-126]. HRP validity may be further restricted by specific applications.
```
bech32::Encode rejects uppercase letters.

Author: l0rinc

src/test/fuzz/bech32.cpp

@@ -4,41 +4,66 @@
 
 #include <bech32.h>
 #include <test/fuzz/fuzz.h>
+#include <test/fuzz/FuzzedDataProvider.h>
 #include <test/util/str.h>
 #include <util/strencodings.h>
 
 #include <cassert>
 #include <cstdint>
 #include <string>
-#include <utility>
 #include <vector>
 
-FUZZ_TARGET(bech32)
+FUZZ_TARGET(bech32_random_decode)
 {
-    const std::string random_string(buffer.begin(), buffer.end());
-    const auto r1 = bech32::Decode(random_string);
-    if (r1.hrp.empty()) {
-        assert(r1.encoding == bech32::Encoding::INVALID);
-        assert(r1.data.empty());
+    auto limit = bech32::CharLimit::BECH32;
+    FuzzedDataProvider fdp(buffer.data(), buffer.size());
+    auto random_string = fdp.ConsumeRandomLengthString(limit + 1);
+    auto decoded = bech32::Decode(random_string, limit);
+
+    if (decoded.hrp.empty()) {
+        assert(decoded.encoding == bech32::Encoding::INVALID);
+        assert(decoded.data.empty());
     } else {
-        assert(r1.encoding != bech32::Encoding::INVALID);
-        const std::string reencoded = bech32::Encode(r1.encoding, r1.hrp, r1.data);
+        assert(decoded.encoding != bech32::Encoding::INVALID);
+        auto reencoded = bech32::Encode(decoded.encoding, decoded.hrp, decoded.data);
         assert(CaseInsensitiveEqual(random_string, reencoded));
     }
+}
+
+// https://github.com/bitcoin/bips/blob/master/bip-0173.mediawiki and https://github.com/bitcoin/bips/blob/master/bip-0350.mediawiki
+std::string GenerateRandomHRP(FuzzedDataProvider& fdp)
+{
+    std::string hrp;
+    size_t length = fdp.ConsumeIntegralInRange<size_t>(1, 83);
+    for (size_t i = 0; i < length; ++i) {
+        // Generate lowercase ASCII characters in ([33-126] - ['A'-'Z']) range
+        char c = fdp.ConsumeBool()
+                 ? fdp.ConsumeIntegralInRange<char>(33, 'A' - 1)
+                 : fdp.ConsumeIntegralInRange<char>('Z' + 1, 126);
+        hrp += c;
+    }
+    return hrp;
+}
+
+FUZZ_TARGET(bech32_roundtrip)
+{
+    FuzzedDataProvider fdp(buffer.data(), buffer.size());
+    auto hrp = GenerateRandomHRP(fdp);
 
-    std::vector<unsigned char> input;
-    ConvertBits<8, 5, true>([&](unsigned char c) { input.push_back(c); }, buffer.begin(), buffer.end());
+    auto input_chars = fdp.ConsumeBytes<unsigned char>(fdp.ConsumeIntegralInRange<size_t>(0, 82));
+    std::vector<uint8_t> converted_input;
+    ConvertBits<8, 5, true>([&](auto c) { converted_input.push_back(c); }, input_chars.begin(), input_chars.end());
 
-    // Input data part + 3 characters for the HRP and separator (bc1) + the checksum characters
-    if (input.size() + 3 + bech32::CHECKSUM_SIZE <= bech32::CharLimit::BECH32) {
-        // If it's possible to encode input in Bech32(m) without exceeding the bech32-character limit:
-        for (auto encoding : {bech32::Encoding::BECH32, bech32::Encoding::BECH32M}) {
-            const std::string encoded = bech32::Encode(encoding, "bc", input);
+    auto size = converted_input.size() + hrp.length() + std::string({bech32::SEPARATOR}).size() + bech32::CHECKSUM_SIZE;
+    if (size <= bech32::CharLimit::BECH32) {
+        for (auto encoding: {bech32::Encoding::BECH32, bech32::Encoding::BECH32M}) {
+            auto encoded = bech32::Encode(encoding, hrp, converted_input);
             assert(!encoded.empty());
-            const auto r2 = bech32::Decode(encoded);
-            assert(r2.encoding == encoding);
-            assert(r2.hrp == "bc");
-            assert(r2.data == input);
+
+            const auto decoded = bech32::Decode(encoded);
+            assert(decoded.encoding == encoding);
+            assert(decoded.hrp == hrp);
+            assert(decoded.data == converted_input);
         }
     }
-}
+}