Should Buddha Bahadur Tamang keep Buddha as part of its given name? (Buddha, Siddha, Gyani, Guru opening Nepali compound given names) #869

Four title words open common Nepali compound given names, and the parser reads them as titles, leaving the given name as the second half of the compound:

>>> parse("Buddha Bahadur Tamang")   # title Buddha, given 'Bahadur', family Tamang
>>> parse("Gyani Maya Gurung")       # title Gyani,  given 'Maya'
>>> parse("Siddha Raj Joshi")        # title Siddha, given 'Raj'
>>> parse("Guru Prasad Adhikari")    # title Guru,   given 'Prasad'
>>> parse("Tamang, Buddha Bahadur")  # title Buddha, given 'Bahadur'

No ambiguity is reported. buddha and siddha are religious titles in en.toml, gyani a religious title in pa.toml (same_as='giani'), and guru the renunciate title of #346 (addresses='given', in hi.toml, bn.toml and pa.toml).

In the nepal-candidates local corpus (#866, fetched 2026-10-24)

These are compound given names (Buddha Bahadur, Gyani Maya, Guru Prasad), not a title before a renounced surname: three or four words, with a family name last. Outside Nepal the sources barely see the words: SSA lists Buddha with 6 boys, and none of the four is in nam_dict.

The question

#346 decided guru is a title that addresses by the given name, for the renunciate class (Guru Nanak, Swami Vivekananda). Here the same word is the first half of a given name. Options:

1. Take buddha, siddha and gyani out of the titles, as Should Mahdi Karimi keep Mahdi as its given name, and Bard Larsen keep Bard? #867 asks for mahdi and bard. The title readings they serve (a Buddha, a Siddha before a name) are rare as written honorifics before a person's name.
2. Keep them, and stand a leading title down when the word after it is a known given-name element (bahadur, maya, prasad). That is a shape change, and would need a list of its own.
3. Leave them, recording the cost in rules/H1's Accepted clause.

guru is the hard one: it is the renunciate title #346 was about, and Guru Prasad is the compound given name.

Related: #867 (the same class for Mahdi, Bard), #346.