[js/webgpu] Fix Expand/Gather when input type is bool (#18999)

### Description
Also update the op test suite.

### Motivation and Context
Previously the *total* size in case `Expand - last dim is not divisible
by 4` was a multiple of 4, even though the *last dimension* was not, so
the bug has never been caught.

js/web/lib/wasm/jsep/webgpu/ops/expand.ts

@@ -47,7 +47,7 @@ const createExpandProgramInfo = (inputs: readonly TensorView[]): ProgramInfo =>
   const outputShape: number[] = calculateOutputShape(inputShape, shape);
   const dataType = inputs[0].dataType;
   const components = dataType === DataType.bool ? 4 : 1;
-  const outputSize = ShapeUtil.size(outputShape) / components;
+  const outputSize = Math.ceil(ShapeUtil.size(outputShape) / components);
 
   const enableInputShapeUniform = enableShapesUniforms(inputShape.length);
   const enableOutputShapeUniform = enableShapesUniforms(outputShape.length);

js/web/lib/wasm/jsep/webgpu/ops/gather.ts

@@ -31,7 +31,7 @@ const createGatherProgramInfo = (inputs: readonly TensorView[], attributes: Gath
 
   const axisDimLimit = inputShape[axis];
   const components = inputs[0].dataType === DataType.bool ? 4 : 1;
-  const outputSize = ShapeUtil.size(outputShape) / components;
+  const outputSize = Math.ceil(ShapeUtil.size(outputShape) / components);
 
   const enableInputShapesUniforms = enableShapesUniforms(inputs[0].dims.length);
   const inputShapeOrRank = enableInputShapesUniforms ? inputs[0].dims.length : inputs[0].dims;

js/web/test/data/ops/expand.jsonc

@@ -168,20 +168,39 @@
         "name": "Expand - last dim is not divisible by 4",
         "inputs": [
           {
-            "data": [true, false, false, true, true, true, false, false, false, true, true, true],
-            "dims": [2, 6],
+            "data": [true, false, false, true, true, true],
+            "dims": [1, 6],
             "type": "bool"
           },
           {
-            "data": [2, 1],
+            "data": [3, 1],
             "dims": [2],
             "type": "int64"
           }
         ],
         "outputs": [
           {
-            "data": [true, false, false, true, true, true, false, false, false, true, true, true],
-            "dims": [2, 6],
+            "data": [
+              true,
+              false,
+              false,
+              true,
+              true,
+              true,
+              true,
+              false,
+              false,
+              true,
+              true,
+              true,
+              true,
+              false,
+              false,
+              true,
+              true,
+              true
+            ],
+            "dims": [3, 6],
             "type": "bool"
           }
         ]

js/web/test/data/ops/gather.jsonc

@@ -99,6 +99,28 @@
     "operator": "Gather",
     "attributes": [],
     "cases": [
+      {
+        "name": "data[4] indices[]",
+        "inputs": [
+          {
+            "data": [false, true, false, false],
+            "dims": [4],
+            "type": "bool"
+          },
+          {
+            "data": [1],
+            "dims": [],
+            "type": "int32"
+          }
+        ],
+        "outputs": [
+          {
+            "data": [true],
+            "dims": [],
+            "type": "bool"
+          }
+        ]
+      },
       {
         "name": "data[2,4] indices[1]",
         "inputs": [