CPSPI::LockToOrigin is used for locking a process to documents from a given set, but this varies in granularity.  With Site Isolation's SitePerProcess policy enabled, (almost) all processes are locked to a site (i.e., scheme + eTLD+1, such as https://google.com).  However, it's also possible to lock some processes to an origin (e.g., https://accounts.google.com) with IsolateOrigins, while other times a process may be locked to a broader category (e.g., file://, or perhaps chrome-extension:// in the future).  Thus, calling it LockToOrigin is confusing and inaccurate.

To eliminate confusion, we should rename it to LockToPrincipal, and eventually start moving more of the "site" terminology toward "principal" in general when it doesn't actually refer to a site (e.g., perhaps SiteInstance).

This came up in https://chromium-review.googlesource.com/c/chromium/src/+/1012502, where the actual "site" concept was important (vs cases where the lock might be broader than that).