fix(build): restore C++ include order broken by clang-format include sort

The clang-format enforcement commit ran with SortIncludes: CaseSensitive, which
alphabetically reordered includes and moved jni_helpers.hpp / json_helpers.hpp
ahead of the server-*.h headers that define the `json` alias they depend on. CI
then failed to build jllama with "'json' does not name a type" cascading through
json_helpers.hpp / jni_helpers.hpp / jllama.cpp on the manylinux, Linux aarch64,
Android and C++-test compilers. (A local build masks it — the local toolchain
resolves `json` regardless of include order.)

Set SortIncludes: Never in .clang-format (the project has order-sensitive includes,
documented in CLAUDE.md) and restore the required order — server-*.h + utils.hpp
before the helper headers — in jllama.cpp and the affected C++ test files. Document
the constraint in CLAUDE.md. clang-format --dry-run --Werror stays clean; the
affected targets rebuild and the C++ tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014L2dLbAtwdq7C6a2gFRsQQ

Author: claude

.clang-format

@@ -175,7 +175,10 @@ RequiresClausePosition: OwnLine
 RequiresExpressionIndentation: OuterScope
 SeparateDefinitionBlocks: Leave
 ShortNamespaceLines: 1
-SortIncludes:    CaseSensitive
+# Never reorder #include lines: this project has order-sensitive includes — the upstream
+# server-*.h headers must precede json_helpers.hpp / jni_helpers.hpp (which use the `json`
+# alias those headers define). Alphabetical sorting breaks the build (json undefined).
+SortIncludes:    Never
 SortJavaStaticImport: Before
 SortUsingDeclarations: LexicographicNumeric
 SpaceAfterCStyleCast: false

CLAUDE.md

@@ -407,6 +407,13 @@ The generated JNI header `src/main/cpp/jllama.h` (produced by `javac -h`) is int
 To bump the enforced version, update the pin in **both** the workflow (`CLANG_FORMAT_VERSION`) and
 this line, then reformat the whole tree with the new version in the same commit.
 
+**`.clang-format` sets `SortIncludes: Never` — do not re-enable include sorting.** The project has
+order-sensitive includes (see the "Include order rule" above): the upstream `server-*.h` headers and
+`utils.hpp` must precede `json_helpers.hpp` / `jni_helpers.hpp`, which use the `json` alias those
+headers define. Alphabetical sorting moves the helper headers first and breaks the build with
+`'json' does not name a type` (it slips past a local build whose toolchain resolves `json` anyway,
+but fails the manylinux/aarch64/Android CI compilers). Keep include order manual.
+
 ### Javadoc — must build cleanly before `mvn package`
 
 The release packaging job runs `mvn package` with the `release` profile, which attaches

src/main/cpp/jllama.cpp

@@ -7,18 +7,18 @@
 
 #include "arg.h"
 #include "build-info.h"
-#include "jni_helpers.hpp"
 #include "json-schema-to-grammar.h"
 #include "llama.h"
 #include "log.h"
-#include "log_helpers.hpp"
 #include "nlohmann/json.hpp"
-#include "server-chat.h"
-#include "server-common.h"
 #include "server-context.h"
 #include "server-queue.h"
 #include "server-task.h"
+#include "server-common.h"
+#include "server-chat.h"
 #include "utils.hpp"
+#include "jni_helpers.hpp"
+#include "log_helpers.hpp"
 
 #include <atomic>
 #include <chrono>

src/test/cpp/test_jni_helpers.cpp

@@ -22,13 +22,13 @@
 
 #include <gtest/gtest.h>
 
-#include "jni_helpers.hpp"
-#include "server-chat.h"
-#include "server-common.h"
 #include "server-context.h"
 #include "server-queue.h"
 #include "server-task.h"
+#include "server-common.h"
+#include "server-chat.h"
 #include "utils.hpp"
+#include "jni_helpers.hpp"
 #include <cstring>
 #include <memory>
 #include <string>

src/test/cpp/test_json_helpers.cpp

@@ -28,13 +28,13 @@
 #include <string>
 #include <vector>
 
-#include "json_helpers.hpp"
-#include "server-chat.h"
-#include "server-common.h"
 #include "server-context.h"
 #include "server-queue.h"
 #include "server-task.h"
+#include "server-common.h"
+#include "server-chat.h"
 #include "utils.hpp"
+#include "json_helpers.hpp"
 
 // ============================================================
 // Minimal fake result types

src/test/cpp/test_server.cpp

@@ -20,11 +20,11 @@
 
 #include <gtest/gtest.h>
 
-#include "server-chat.h"
-#include "server-common.h"
 #include "server-context.h"
 #include "server-queue.h"
 #include "server-task.h"
+#include "server-common.h"
+#include "server-chat.h"
 #include "utils.hpp"
 
 // ============================================================