The first verse of the Bṛhatkathā?

This verse from the Gāthāmuktāvalī (a collection of Prakrit verses based on the Sattasaī) is cited in at least three other places. It has been identified as the first verse (maṅgalācaraṇa) of the lost Bṛhatkathā of Guṇāḍhya. I give the other versions with some notes below.

पनमत पनअ-प्पकुपित-कोली-चलनग्ग-लग्ग-पटिबिंबं ।
तससु नख-तप्पनेसुं एकातस-तनु-थलं लुद्दं ॥

(Gāthāmuktāvalī 13)

Reverence to Rudra, whose image clings to the toes of Gaurī in her flirtatious anger, who takes eleven forms among the ten mirrors of her nails.

Hemacandra knows this verse (probably from Bhoja) and cites it as an example of the optional change of r into l in cūlikapaiśācika (a special form of the paiśācika language) in Siddhahemacandra 8.4.326:
पनमथ पनय-पकुप्पित-गोली-चलनग्ग-लग्ग-पतिबिंबं ।
तससु नखतप्पनेसुं एकातस-तनु-थलं लुद्दं ॥

Bhoja cites this verse twice. In Śṛṅgāraprakāśa [ed. Dwivedī] 3.61, he cites it as an example of paiśāca, one of two kinds of śliṣṭa Prakrit.
पनमथ पनय-पकुप्पित-को/गोली-चलनग्ग-लग्ग-पतिबिंबं ।
तससु नख-तप्पनेसुं एकातस तनु-थलं लुद्दं/रुत्तं ॥

In Sarasvatīkaṇṭhābharaṇa 2 ex. 4, he cites it as an example of paiśācī, one of the six pure (śuddha) kinds of language:
पनमत पनअपकुप्पितगोलीचलनग्गलग्गपडिबिम्बम् ।
दससु नहदप्पनेसु एआदसतनुधलं लुद्दम् ॥

Notes:

• Context-free changes:
• No contrast between n and ṇ. (Namisādhu नणोर्नकारः; Hemacandra 8.4.306 = ps.-Vararuci 10.5 णो नः)
• No contrast between s, ś and s. (Hemacandra 8.4.309 श-षोः सः)
• No contrast between l and r. (Hemacandra 8.4.326 रस्य लो वा, but compare the regular Paiśācika rule 8.4.308 लो ळः).
• Intervocalic consonants:
• Mostly preserved as in Sanskrit, with assimilation. पनअ॰ for पनय॰, नह॰ for नख॰, and एआ॰ for एका॰, are Prakritisms in the Sarasvatīkaṇṭhābharaṇa version.
• Voiced geminates are allowed (ludda-, -agga-).
• पडिबिंबं (as in Prakrit) or पटिबिंबं (Hemacandra cites पटिमा at 8.4.325)?
• Initial consonants:
• No contrast between t and d (except in the Sarasvatīkaṇṭhābharaṇa version). (Namisādhu दस्य वा तकारः; Hemacandra 8.4.307 तदोस्तः)
• No contrast between th and dh (except in the Sarasvatīkaṇṭhābharaṇa version).
• Other voiced/unvoiced contrasts are ambiguous. kolī or golī?
• Morphology:
• 2nd. pl. imperative -ta or -tha? The former is the Sanskrit ending, the latter is Prakrit (-ha, -dha in Śaurasenī).
• ppakupita is preferable to pakuppita (< prakupita).

The principal issue is the scope and context for devoicing consonants. For ps.-Vararuci 10.3 (वर्गाणां तृतीयचतुर्थयोरयुजोरनाद्योराद्यौ), this doesn't apply to initial consonants; Hemacandra 8.4.327 (नादि-युज्योरन्येषाम्) hesitates about the exception. Namisādhu only knows the devoicing of t, not a general devoicing rule.

Could this be the maṅgalācaraṇa of the Bṛhatkathā?