My background is with embedded systems as well, but not specifically mobile platforms.
I have not had an opportunity to test with only the base OS and I respect your findings.
My test case is different and is the basis for my post.
I have two identical iPhone 5s (one mine, one my wife's), purchased at the same time, both with iOS 10.3.2 installed, via iTunes, installed the same way, at the same point in time. The phones share a common set of applications installed, but also differ with many user-specific applications, differ in the data associated with the applications, and differ in the usage patterns of the applications.
One of the phones exhibits the battery drain issue, random overheating, and what appears to be the 30% battery shutdown issues. The other does not.
In the past, this same phenomenon has been observed, and after several/many weeks as applications are updated for the new version of iOS the problem disappears. I have not had the time or patience to track the updates to try determine exactly when the issue goes away. Note too that the behavior has not been tied to one specific phone (i.e. sometimes it is mine that has issues, sometimes my wife's phone) so I have ruled out hardware issues as well.
It seems this difference in behavior must be triggered by application software, data, or some environmental difference between the two phones. As I type this, I am pondering if it could be in some obscure configuration area such as carrier settings / roaming (wild speculation as I am not familiar with the deep details of mobile phone operation).
Thank you for your insight.