[bmi2] Add lzcnt and bmi target attributes

* When dynamic dispatching to bmi2 add lzcnt and bmi to the
  TARGET_ATTRIBUTE.
* Centralize the bmi2 TARGET_ATTRIBUTE definition to
  BMI2_TARGET_ATTRIBUTE so we can change it in the future.
* Only enable bmi2 when both bmi1 & bmi2 are supported. There shouldn't
  be any cases where bmi2 is supported but bmi1 isn't. But, since we are
  using the instruction we should check bmi1 as well.

lib/compress/huf_compress.c

@@ -1029,7 +1029,7 @@ HUF_compress1X_usingCTable_internal_body(void* dst, size_t dstSize,
 
 #if DYNAMIC_BMI2
 
-static TARGET_ATTRIBUTE("bmi2") size_t
+static BMI2_TARGET_ATTRIBUTE size_t
 HUF_compress1X_usingCTable_internal_bmi2(void* dst, size_t dstSize,
                                    const void* src, size_t srcSize,
                                    const HUF_CElt* CTable)

lib/compress/zstd_compress.c

@@ -12,7 +12,6 @@
 *  Dependencies
 ***************************************/
 #include "../common/zstd_deps.h"  /* INT_MAX, ZSTD_memset, ZSTD_memcpy */
-#include "../common/cpu.h"
 #include "../common/mem.h"
 #include "hist.h"           /* HIST_countFast_wksp */
 #define FSE_STATIC_LINKING_ONLY   /* FSE_encodeSymbol */
@@ -100,7 +99,7 @@ static void ZSTD_initCCtx(ZSTD_CCtx* cctx, ZSTD_customMem memManager)
     assert(cctx != NULL);
     ZSTD_memset(cctx, 0, sizeof(*cctx));
     cctx->customMem = memManager;
-    cctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid());
+    cctx->bmi2 = ZSTD_cpuSupportsBmi2();
     {   size_t const err = ZSTD_CCtx_reset(cctx, ZSTD_reset_parameters);
         assert(!ZSTD_isError(err));
         (void)err;

lib/compress/zstd_compress_sequences.c

@@ -399,7 +399,7 @@ ZSTD_encodeSequences_default(
 
 #if DYNAMIC_BMI2
 
-static TARGET_ATTRIBUTE("bmi2") size_t
+static BMI2_TARGET_ATTRIBUTE size_t
 ZSTD_encodeSequences_bmi2(
             void* dst, size_t dstCapacity,
             FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,