You might want to read a scholarly introduction to the field (not a pop-sci one that is likely to be unreliable by definition) from recent years.
Each descendant subregion of PIE of course imposed its own religious beliefs and local gods. There is no surprise there. The key thread of derivative localized polytheism, often rooted in the same nature-based phenomenon, however persisted across the regions.
Refer to the sibling comment by drivers99. Sky god is not a vague lexeme; it is a critical rooting evidence. You're fooling no one by attempting to cover it up. A shallow dismissal of valid evidence is an attribute of a conman.
No one has any problem with reconstructing some branch-level stuff like Italic or Indo-Iranian. But projecting it back to even Nuclear/Core Indo-European, let alone Proto-Indo-European, is too fraught with problems.
Where these practices often differed strongly from Christianity was that they were not totalizing, overarching, mandatory systems of belief. They were more about practices embedded in daily and civic life -- sacrifices, festivals, acts of veneration, etc. -- than about belief or internal mental state (“Whosoever believeth in him should not perish, but have everlasting life”). They were also generally quite willing to incorporate gods, heroes, and practices from other peoples encountered along the way.
IMHO, exclusive, universal, doctrinally policed religion developed partly through the interaction of monotheistic traditions with imperial states, which had strong reasons to standardize institutions and identities; although neither Rome nor the caliphates ever achieved simple religious homogeneity.
(And yeah, clearly there's a lot of commonality between many of the cultures derived from early Indo-European migration etc but also incredible diversity and very clear adaptations to wherever they ended up and who they ended up co-existing with)
Yes, they were not not overarching, mandatory systems of belief, but is not an excuse to dismiss them as religions. The religions were real. After all, we see the same continuing religious beliefs that you noted (sacrifices, festivals, veneration, etc.) in Hinduism, which traces a direct chain to the PIE religion.
I don't question your credentials but I also think many things deemed to be IIr innovations may not even be. IIr came through CW -> Sintashta after all and BMAC influences are overstated. For instance, Soma may have a root in this region, but the myth associated with theft of divine intoxicating drink and it's function as the mead of poetry must predate the specific IIr Soma cult.
PIE sóh₂wl̥ (“sun”) → Proto-Germanic sunnǭ (“sun”) → Old English sunne (“sun”) → Old English Sunnandæg (“Sun's day”) → Middle English Sonday / Sunday → Sunday
PIE mḗh₁n̥s (“moon; month”) → Proto-Germanic mēnô (“moon”) → Old English mōna (“moon”) → Old English Mōnandæg (“Moon's day”) → Middle English Monenday / Monday → Monday
PIE deywós → Sanskrit deva, “god” → Latin deus, “god” → Old Norse Týr, the god → Old English Tīw → Tīwesdæg (“Tiw's day”) → Tuesday
PIE dyew- (“bright sky; daylight”) → deywós (“celestial one; god”) → Proto-Germanic Tīwaz → Old English Tīw → Tīwesdæg → Tuesday
The reconstruction exists for the remaining days too, but is more uncertain when tracing it to PIE:
PIE weh₂t- / related roots concerning excitement or inspiration → Proto-Germanic *Wōðanaz → Old English Wōden → Wōdnesdæg (“Woden's day”) → Wednesday
PIE (s)tenh₂- (“thunder”) (reconstruction/derivation has some uncertainty) → Proto-Germanic þunraz (“thunder”; the god Thunder/Thor) → Old English Þunor → Þunresdæg (“Þunor's day”) → Thursday
PIE prih₁- / preyH- (“to love; be dear”) → Proto-Germanic Frijjō (the goddess Frigg, approximately “beloved/dear one”) → Old English Frīg → Old English Frīgedæg (“Frīg's day”) → Middle English Fridai / Friday → Friday
Saturday is not known to trace to PIE:
Latin Saturnus → Old English Sætern(es)dæg (“Saturn's day”) → Saturday
There can be and there often is correlation, but it's not the same thing. They don't have to move together.
It is mathematically true that they don't have to move together, but there are lines of evidence that for PIE they did. For example, there are lots of divinity related words and root words found in the PIE language that then carried forward in various derived languages in distinct geographical regions of Indo-Europe. This is not nothing.
There are even be stories that similarly exist in common. There is the story of slaying the serpent. There is another story of the divine twins. These stories are found everywhere that the PIE culture went.
The third line of evidence is PIE poetic formulas and poetic traditions. These again show up in multiple post-PIE cultures. The local writings show the same mythic structure across cultures. These things are too difficult to explain without a common ancestor culture/religion.