Fix jint->size_t integer narrowing in DecoderJNI and EncoderJNI nativePush

[momo-123-coder]

Fix jint→size_t integer narrowing in DecoderJNI and EncoderJNI nativePush

Summary

nativePush in both decoder_jni.cc and encoder_jni.cc accepts
jint input_length (a signed 32-bit JNI integer) and assigns it directly
to a size_t field without a sign check:

handle->input_length = input_length;   // decoder_jni.cc:133
handle->input_last   = input_length;   // encoder_jni.cc:147

When the Java caller passes a negative value (e.g., -1, or any value
where the high bit is set), the implicit jint → size_t conversion wraps
to SIZE_MAX - |value| + 1 (4 294 967 295 for -1 on 32-bit, or
18 446 744 073 709 551 615 on 64-bit). This causes:

• in_size = handle->input_length - handle->input_offset → huge value
• BrotliDecoderDecompressStream is called with a multi-gigabyte declared
  input size pointing at attacker-controlled memory
• On instrumented builds: heap-buffer-overflow / process abort
• On production: out-of-bounds read, potential information disclosure,
  or remote code execution depending on heap layout

Root cause

jint is defined as int32_t in <jni.h>. The JNI specification allows
any Java int value, including negative values, to reach native code.
There is no existing guard against input_length < 0 before the assignment.

Fix

Add a sign check and replace the implicit narrowing cast with an explicit
static_cast<size_t> in both files:

if (input_length < 0) {
  return;
}
handle->input_offset = 0;
handle->input_length = static_cast<size_t>(input_length);

The encoder_jni.cc path is fixed identically (input_last field).

Testing

Existing unit tests continue to pass. A LibFuzzer harness that exercises
this path with negative input_length values is available on request.

Security impact

• CWE-191: Integer Underflow (Wrap or Wraparound)
• CWE-125: Out-of-Bounds Read
• CVSS 3.1: AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H — 9.1 CRITICAL
  (network-reachable through any JVM processing untrusted compressed data)

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.