ggml: add Q1_0 1-bit quantization support (CPU) (#21273)

* ggml: add Q1_0 and Q1_0_g128 1-bit quantization support (CPU)

* add generic fallback for x86

* remove Q1_0 (group size 32)

* rename Q1_0_g128 => Q1_0

* fix Q1_0 LlamaFileType Enum

* Fix trailing spaces; add generic fallback for othre backends

* Apply suggestions from code review

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* fix /r/n spacing + arch-fallback

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

ggml/src/ggml-common.h

@@ -93,6 +93,10 @@ typedef sycl::half2 ggml_half2;
 // QR = QK / number of values before dequantization
 // QI = number of 32 bit integers before dequantization
 
+#define QI1_0 (QK1_0 / 32)
+#define QR1_0 1
+
+
 #define QI4_0 (QK4_0 / (4 * QR4_0))
 #define QR4_0 2
 
@@ -170,6 +174,13 @@ typedef sycl::half2 ggml_half2;
 #define GGML_EXTENSION __extension__
 #endif // _MSC_VER
 
+#define QK1_0 128
+typedef struct {
+    ggml_half d;           // delta
+    uint8_t qs[QK1_0 / 8]; // bits / quants
+} block_q1_0;
+static_assert(sizeof(block_q1_0) == sizeof(ggml_half) + QK1_0 / 8, "wrong q1_0 block size/padding");
+
 #define QK4_0 32
 typedef struct {
     ggml_half d;           // delta