The examples in GG5 and prog also use the kanji form.

Google N-gram Corpus Counts
╭─ーーーーー─┬────────┬───────╮
│ 　粋がっ　 │ 13,812 │ 30.4% │
│ 　意気がっ │  1,917 │  4.2% │
│ 　いきがっ │ 24,779 │ 54.5% │
│ 　イキがっ │  4,977 │ 10.9% │
├─ーーーーー─┼────────┼───────┤
│ に粋がっ　 │    770 │ 30.3% │
│ に意気がっ │    136 │  5.4% │
│ にいきがっ │  1,341 │ 52.8% │
│ にイキがっ │    291 │ 11.5% │
╰─ーーーーー─┴────────┴───────╯

Could add a [uk] tag. I see that sankoku doesn't use its "okay to write in kana" braces for this word, though.

@@ -15 +15 @@
-<re_nokanji/>
+<re_inf>&sk;</re_inf>

Google N-gram Corpus Counts
╭─ーーーー─┬───────┬───────╮
│ 粋がる　 │ 4,069 │ 62.7% │
│ 意気がる │ 1,033 │ 15.9% │
│ イキがる │ 1,384 │ 21.3% │
╰─ーーーー─┴───────┴───────╯

@@ -11,0 +12,4 @@
+</r_ele>
+<r_ele>
+<reb>イキがる</reb>
+<re_nokanji/>

Daijirin and nikkoku only have a yodan (v4r) part-of-speech tag in their entries for すいがる. They reference works from the late 1700s / early 1800s for the word. I don't think it can be merged here and it's not worth making a new entry for it.

https://github.com/JMdictProject/JMdictIssues/issues/103#issuecomment-1836978316

There are many entries in daijirin that begin with reading notes like 〔古くは「〇〇」〕. Generally the readings are pretty similar to the modern reading and only differ by a dakuten or vowel sound. I think those are the sorts of readings we want to use with the [ok] tag now.

@@ -12,5 +11,0 @@
-</r_ele>
-<r_ele>
-<reb>すいがる</reb>
-<re_restr>粋がる</re_restr>
-<re_inf>&ok;</re_inf>

daijr/s, nikk

@@ -11,0 +12,5 @@
+</r_ele>
+<r_ele>
+<reb>すいがる</reb>
+<re_restr>粋がる</re_restr>
+<re_inf>&ok;</re_inf>