getElementsByTagName

Requirements

• Export getElementsByTagName(root, tagName) as named and default. root is an Element, Document, or DocumentFragment. Return an Element[] snapshot, excluding root itself.
• Match HTML tag names case-insensitively. A tagName of * matches all descendant elements. Preserve depth-first document order.
• Visit only elements as matches; text and comment nodes do not appear in the result. Support detached trees and empty roots.
• Implement the traversal yourself. Do not call native getElementsByTagName, querySelector, querySelectorAll, TreeWalker, or NodeIterator.
• Do not mutate the tree. Later DOM changes must not change an already returned array. Invalid tag names and namespace-specific matching are outside scope.

Examples

Input: getElementsByTagName(container, "BUTTON")
Output: An array of all descendant button elements in document order.

Input: getElementsByTagName(container, "*")
Output: All descendant elements, excluding container itself.