[JS IR] stdlib: port methods that use bit representation of numbers

- Port part of 'misc.js' from current backend to 'bitUtils.kt' in IR backend
- Enable toBits, toRawBits, fromBits extension function on Double and Float
- Enable Double.withSign
- Refactor getNumberHashCode using new utils

libraries/stdlib/js-ir/src/kotlin/math.kt

@@ -0,0 +1,19 @@
+/*
+ * Copyright 2010-2018 JetBrains s.r.o. and Kotlin Programming Language contributors.
+ * Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
+ */
+package kotlin.math
+
+import kotlin.js.internal.BitUtils
+
+/**
+ * Returns this value with the sign bit same as of the [sign] value.
+ *
+ * If [sign] is `NaN` the sign of the result is undefined.
+ */
+@SinceKotlin("1.2")
+public actual fun Double.withSign(sign: Double): Double {
+    val thisSignBit = BitUtils.doubleSignBit(this)
+    val newSignBit = BitUtils.doubleSignBit(sign)
+    return if (thisSignBit == newSignBit) this else -this
+}

libraries/stdlib/js-ir/src/kotlin/numbers.kt

@@ -0,0 +1,64 @@
+/*
+ * Copyright 2010-2018 JetBrains s.r.o. and Kotlin Programming Language contributors.
+ * Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
+ */
+
+package kotlin
+
+import kotlin.js.internal.BitUtils
+
+/**
+ * Returns a bit representation of the specified floating-point value as [Long]
+ * according to the IEEE 754 floating-point "double format" bit layout.
+ */
+@SinceKotlin("1.2")
+public actual fun Double.toBits(): Long =
+    BitUtils.doubleToRawBits(if (this.isNaN()) Double.NaN else this)
+
+/**
+ * Returns a bit representation of the specified floating-point value as [Long]
+ * according to the IEEE 754 floating-point "double format" bit layout,
+ * preserving `NaN` values exact layout.
+ */
+@SinceKotlin("1.2")
+public actual fun Double.toRawBits(): Long =
+    BitUtils.doubleToRawBits(this)
+
+/**
+ * Returns the [Double] value corresponding to a given bit representation.
+ */
+@SinceKotlin("1.2")
+@kotlin.internal.InlineOnly
+public actual inline fun Double.Companion.fromBits(bits: Long): Double =
+    BitUtils.doubleFromBits(bits)
+
+/**
+ * Returns a bit representation of the specified floating-point value as [Int]
+ * according to the IEEE 754 floating-point "single format" bit layout.
+ *
+ * Note that in Kotlin/JS [Float] range is wider than "single format" bit layout can represent,
+ * so some [Float] values may overflow, underflow or loose their accuracy after conversion to bits and back.
+ */
+@SinceKotlin("1.2")
+public actual fun Float.toBits(): Int =
+    BitUtils.floatToRawBits(if (this.isNaN()) Float.NaN else this)
+
+/**
+ * Returns a bit representation of the specified floating-point value as [Int]
+ * according to the IEEE 754 floating-point "single format" bit layout,
+ * preserving `NaN` values exact layout.
+ *
+ * Note that in Kotlin/JS [Float] range is wider than "single format" bit layout can represent,
+ * so some [Float] values may overflow, underflow or loose their accuracy after conversion to bits and back.
+ */
+@SinceKotlin("1.2")
+public actual fun Float.toRawBits(): Int =
+    BitUtils.floatToRawBits(this)
+
+/**
+ * Returns the [Float] value corresponding to a given bit representation.
+ */
+@SinceKotlin("1.2")
+@kotlin.internal.InlineOnly
+public actual inline fun Float.Companion.fromBits(bits: Int): Float =
+    BitUtils.floatFromBits(bits)