feat: list all versions from spark releases page

Note: the versions here are different from the downloads and archive
pages, so this will have to change

lib/utils.bash

@@ -24,16 +24,13 @@ sort_versions() {
 		LC_ALL=C sort -t. -k 1,1 -k 2,2n -k 3,3n -k 4,4n -k 5,5n | awk '{print $2}'
 }
 
-list_github_tags() {
-	git ls-remote --tags --refs "$GH_REPO" |
-		grep -o 'refs/tags/.*' | cut -d/ -f3- |
-		sed 's/^v//' # NOTE: You might want to adapt this sed to remove non-version strings from tags
-}
-
 list_all_versions() {
-	# TODO: Adapt this. By default we simply list the tag names from GitHub releases.
-	# Change this function if spark has other means of determining installable versions.
-	list_github_tags
+	curl -s https://spark.apache.org/releases/ |
+		grep -oE "spark-release(-\d+){3}.html" |
+		grep -oE "spark-release-[0-9]*-[0-9]*-[0-9]*.html" |
+		uniq |
+		sed 's/spark-release-\(.*\).html/\1/g' |
+		sed 's/-/./g'
 }
 
 download_release() {