ICU-23365 Initialize pointer members in TransliterationRule constructor to prevent crash

[OwenSanzas]

Description

TransliterationRule's main constructor (rbt_rule.cpp:58) does not initialize
its pointer members (anteContext, key, postContext, output) in the initializer
list. If the constructor returns early (e.g., when U_FAILURE(status) at line 72),
these pointers remain uninitialized. When the LocalPointer destructor subsequently
calls ~TransliterationRule(), it executes delete on garbage pointer values,
causing a SEGV.

Triggered via the public API Transliterator::createFromRules() with a 15-byte
malformed input.

Root Cause

The initializer list only initializes segments(nullptr) and data(theData).
On early return, anteContext, key, postContext, and output contain garbage.
The destructor unconditionally deletes all of them.

Fix

Add nullptr initialization for all pointer members and zero-initialization for
scalar members in the initializer list. After this fix, early returns leave all
pointers as nullptr, and delete nullptr is a safe no-op per the C++ standard.

Reproduction

Crash input (15 bytes):
```bash
echo 'AAAcAAABABgYJAAYGCsA' | base64 -d > crash.bin
```

Fuzzer:
```cpp
#include <stddef.h>
#include <stdint.h>
#include
#include
#include "unicode/translit.h"
#include "unicode/unistr.h"
#include "unicode/parseerr.h"
#include "unicode/utypes.h"

extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
if (size < 3) return 0;
uint8_t dir = data[0] & 1;
data++; size--;
size_t unistr_size = size / 2;
std::unique_ptr<char16_t[]> fuzzbuff(new char16_t[unistr_size]);
std::memcpy(fuzzbuff.get(), data, unistr_size * 2);
icu::UnicodeString fuzzstr(false, fuzzbuff.get(), unistr_size);

UErrorCode status = U_ZERO_ERROR;
UParseError pe;
std::unique_ptr<icu::Transliterator> t(
    icu::Transliterator::createFromRules(
        UNICODE_STRING_SIMPLE(\"fuzz\"), fuzzstr,
        dir ? UTRANS_FORWARD : UTRANS_REVERSE, pe, status));
if (U_SUCCESS(status) && t) {
    icu::UnicodeString sample(u\"Hello World 123\");
    t->transliterate(sample);
}
return 0;

}
```

Standalone (no fuzzer):
```cpp
#include "unicode/translit.h"
#include "unicode/unistr.h"
#include "unicode/parseerr.h"
#include "unicode/utypes.h"
#include
#include
#include

int main(int argc, char* argv[]) {
std::ifstream file(argv[1], std::ios::binary);
std::vector<uint8_t> data((std::istreambuf_iterator(file)),
std::istreambuf_iterator());
if (data.size() < 3) return 1;
uint8_t dir = data[0] & 1;
size_t unistr_size = (data.size() - 1) / 2;
std::vector<char16_t> buf(unistr_size);
memcpy(buf.data(), data.data() + 1, unistr_size * 2);
icu::UnicodeString rules(false, buf.data(), unistr_size);
UErrorCode status = U_ZERO_ERROR;
UParseError pe;
icu::Transliterator* t = icu::Transliterator::createFromRules(
UNICODE_STRING_SIMPLE("test"), rules,
dir ? UTRANS_FORWARD : UTRANS_REVERSE, pe, status);
if (U_SUCCESS(status) && t) {
icu::UnicodeString sample(u"Hello");
t->transliterate(sample);
delete t;
}
return 0;
}
```

Before fix: SEGV at rbt_rule.cpp:196
After fix: createFromRules failed: U_MEMORY_ALLOCATION_ERROR (graceful error return)

Contribution: Found by FuzzingBrain (O2Lab, Texas A&M University) using a custom transliterator fuzzer targeting the previously unfuzzed createFromRules() API surface.

Checklist

• Required: Issue filed: ICU-23365
• Required: The PR title must be prefixed with a JIRA Issue number
• Required: Each commit message must be prefixed with a JIRA Issue number
• Issue accepted (done by Technical Committee after discussion)
• Tests included, if applicable
• API docs and/or User Guide docs changed or added, if applicable