[Grammar] Migrate to XGrammar (#3015)

This PR migrates the implementation of the grammar engine to
an external library XGrammar. The implementation of XGrammar is
basically the same as our current version, but more modularized and
robust. XGrammar will be continuously enhanced in the future.

See https://github.com/mlc-ai/xgrammar for more details.

.gitmodules

@@ -13,3 +13,6 @@
 [submodule "3rdparty/stb"]
 	path = 3rdparty/stb
 	url = https://github.com/nothings/stb.git
+[submodule "3rdparty/xgrammar"]
+	path = 3rdparty/xgrammar
+	url = https://github.com/mlc-ai/xgrammar.git

CMakeLists.txt

@@ -68,8 +68,11 @@ set(MLC_LLM_RUNTIME_LINKER_LIB "")
 set(TOKENZIER_CPP_PATH 3rdparty/tokenizers-cpp)
 add_subdirectory(${TOKENZIER_CPP_PATH} tokenizers EXCLUDE_FROM_ALL)
 
-
+set(XGRAMMAR_PATH 3rdparty/xgrammar)
 tvm_file_glob(GLOB_RECURSE MLC_LLM_SRCS cpp/*.cc)
+tvm_file_glob(GLOB_RECURSE XGRAMMAR_SRCS ${XGRAMMAR_PATH}/cpp/*.cc)
+list(FILTER XGRAMMAR_SRCS EXCLUDE REGEX "${XGRAMMAR_PATH}/cpp/pybind/.*\\.cc")
+list(APPEND MLC_LLM_SRCS ${XGRAMMAR_SRCS})
 add_library(mlc_llm_objs OBJECT ${MLC_LLM_SRCS})
 
 set(
@@ -83,12 +86,14 @@ set(
 set(MLC_LLM_COMPILE_DEFS ${MLC_LLM_COMPILE_DEFS} DMLC_USE_LOGGING_LIBRARY=<tvm/runtime/logging.h>)
 set(MLC_LLM_COMPILE_DEFS ${MLC_LLM_COMPILE_DEFS} __STDC_FORMAT_MACROS=1)
 set(MLC_LLM_COMPILE_DEFS ${MLC_LLM_COMPILE_DEFS} PICOJSON_USE_INT64)
+set(MLC_LLM_COMPILE_DEFS ${MLC_LLM_COMPILE_DEFS} XGRAMMAR_ENABLE_LOG_DEBUG=0)
 
-target_include_directories(mlc_llm_objs PRIVATE ${MLC_LLM_INCLUDES})
 target_compile_definitions(mlc_llm_objs PRIVATE ${MLC_LLM_COMPILE_DEFS})
-target_include_directories(mlc_llm_objs PRIVATE ${TOKENZIER_CPP_PATH}/include)
 target_compile_definitions(mlc_llm_objs PRIVATE -DMLC_LLM_EXPORTS)
+target_include_directories(mlc_llm_objs PRIVATE ${MLC_LLM_INCLUDES})
 target_include_directories(mlc_llm_objs PRIVATE 3rdparty/stb)
+target_include_directories(mlc_llm_objs PRIVATE ${TOKENZIER_CPP_PATH}/include)
+target_include_directories(mlc_llm_objs PRIVATE ${XGRAMMAR_PATH}/include)
 
 add_library(mlc_llm SHARED $<TARGET_OBJECTS:mlc_llm_objs>)
 add_library(mlc_llm_static STATIC $<TARGET_OBJECTS:mlc_llm_objs>)
@@ -135,7 +140,6 @@ add_library(mlc_llm_module SHARED $<TARGET_OBJECTS:mlc_llm_objs>)
 target_link_libraries(mlc_llm_module PUBLIC tvm)
 target_link_libraries(mlc_llm_module PRIVATE tokenizers_cpp)
 
-
 set_property(TARGET mlc_llm_module APPEND PROPERTY LINK_OPTIONS "${MLC_VISIBILITY_FLAG}")
 set_property(TARGET mlc_llm APPEND PROPERTY LINK_OPTIONS "${MLC_VISIBILITY_FLAG}")