docs(numpy): Add ndarray.byteswap() term entry

Signed-off-by: Prakhar Sharma <prakharsharma_mc24b06_001@dtu.ac.in>

Author: prakhar14-op

content/numpy/concepts/ndarray/terms/byteswap/byteswap.md

@@ -0,0 +1,83 @@
+<!-- ---
+title: byteswap()
+description: "Converts the endianness of an array's data items, optionally in place."
+keywords: 
+  - numpy
+  - ndarray
+  - byteswap
+  - endianness
+  - byte order
+type: "function"
+---
+
+## Description
+
+The **`.byteswap()`** method reverses the byte order (or **endianness**) of the array’s data elements. This process essentially swaps the position of bytes within each multi-byte data item (like integers, floats, or complex numbers).
+
+This feature is primarily used when data must be moved between computer systems that use different byte ordering conventions: 
+
+* **Little-Endian (LE):** Stores the least significant byte first (e.g., Intel/AMD).
+* **Big-Endian (BE):** Stores the most significant byte first (e.g., older network protocols).
+
+The operation is performed **in place** if the `inplace` parameter is set to `True`, modifying the original array.
+
+---
+
+## Syntax
+
+```python
+ndarray.byteswap(inplace=False) -->
+
+
+---
+# REQUIRED METADATA (YAML FRONT MATTER)
+title: numpy.ndarray.byteswap       # Specific title
+description: Modifies the byte order (endianness) of the array elements in place or returns a copy with swapped bytes. # Required Description
+keywords: numpy, ndarray, byteswap, endian, endianness, byte order, data representation # Relevant keywords
+type: method                     # Correct entry type
+---
+
+The **`.byteswap()`** method reverses the byte order (endianness) of each element in a NumPy array. It's used to ensure data compatibility between systems with different byte storage conventions (little-endian vs. big-endian).
+
+---
+## Syntax 
+
+```python
+array_instance.byteswap(inplace=False)
+
+### Parameters
+Parameter,Type,Description,Default
+inplace,bool,"If True, the byte swap happens directly on the original array (in place), and the method returns the array itself. If False, a new array with swapped bytes is created and returned, leaving the original unchanged.",False
+
+### Return Value
+Returns either the modified original array (if inplace=True) or a new array with the byte order swapped (if inplace=False). The data type (dtype) of the returned array may reflect the new byte order depending on the system and NumPy version, but the underlying byte sequence is always swapped.
+
+## Example
+Consider an array of 16-bit integers. On most systems (like Intel/AMD), these are stored in little-endian format (`<i2`). The number 256 requires two bytes; in little-endian, it's `00 01`. Swapping the bytes makes it `01 00`, which represents a different number if interpreted using the original data type's byte order.
+
+## Codebyte
+import numpy as np
+
+# Create an array of 16-bit integers (usually little-endian by default)
+# 256 in little-endian bytes: 00 01
+original_array = np.array([1, 256, 512], dtype=np.int16)
+print(f"Original array: {original_array}")
+print(f"Original dtype: {original_array.dtype}")
+print(f"Original bytes (hex): {original_array.tobytes().hex()}")
+
+# Swap bytes, returning a NEW array (inplace=False is default)
+swapped_array = original_array.byteswap()
+
+print("\n--- After byteswap() ---")
+# Values change because the raw bytes are swapped but interpreted via original dtype order
+print(f"Swapped array values (interpreted): {swapped_array}")
+print(f"Swapped array dtype: {swapped_array.dtype}")
+print(f"Swapped bytes (hex): {swapped_array.tobytes().hex()}") # Bytes reversed per element (e.g., 0001 -> 0100)
+
+# Swap bytes IN PLACE
+print("\n--- In-place swap ---")
+# The method returns the modified array when inplace=True
+modified_original = original_array.byteswap(inplace=True)
+print(f"Original array after in-place swap: {original_array}")
+print(f"Return value (inplace=True): {modified_original}") # It's the same array object
+print(f"Original array bytes after swap (hex): {original_array.tobytes().hex()}")