The idea that the Yogasūtra (henceforth YS) and the Yogabhāṣya (henceforth YBh) are not two distinct texts has been discussed for the first way in a systematic way by Johannes Bronkhorst in 1985 (“Patañjali and the Yoga Sūtras”, Studien zur Indologie und Iranistik). Philipp Maas in his published PhD thesis (Maas 2006) examined it again and Philipp Maas in his contribution to Eli Franco’s Periodization and Historiography of Indian Philosophy (2013) dealt with it again in greater detail.
Bronkhorst suggested that the Yogasūtras have been assembled by the author of the Bhāṣya (“the sūtras were brought together by the author of the Yogabhāṣya“, p. 17), who might have added further sūtras to the lore of transmitted ones and mentions among his evidences the fact that the YS 1.21–23 have an unforced interpretation which has been violated by the author of the Bhāṣya (who, then, evidently took pre-existing sūtras and provided them with a new context and a new interpretation). Accordingly, Bronkhorst suggests that “the available evidence points to two persons, Patañjali [as author of the YS] and Vindhyavāsin [as author of the Bhāṣya]” (p. 18). A large part of Bronkhorst 1985, in fact, points to the reconstruction of the theoretical background of YS and YBh and connects it with Sāṅkhya teachings.
Maas, seems to push the thesis further and maintains that the only text whose existence can be reconstructed is the Pātañjala Yogaśāstra (henceforth PYŚ, as in Maas 2013), already including sūtra and bhāṣya (to be understood not as “different literary genres but compositional elements of scholarly works (śāstra)”, p. 65). In the case of YS 1.2, 1.41 and 2.23, Maas suggests that the fact that they are introduced with the perfect tense pravavṛte instead of the usual present passive might be a hint of the fact that these sūtras were older. Noteworthy is also the fact that Maas uses a different set of arguments than the ones used by Bronkhorst 1985. In harmony with his textual critical interests, Maas picks out manuscript evidences, such as the lack of an independent transmission of the YS, which are only transmitted together with the YBh, the lack of a consistent marking of the sūtras in the manuscripts, and the colophons, which are not present separately for the YS part and which mention the “YBh” of Vyāsa only “in a few manuscripts of limited stemmatic relevance” (p. 58).
A very interesting way to validate Maas’ arguments would be, thus, to test them against the evidence of the other philosophical sūtras. Are not they also only transmitted within their Bhāṣya? And how do their colophons look like? Are the sūtras marked in manuscripts? I only know a little bit about the Mīmāṃsā– and Nyāyasūtra and have no information about their colophons.