Introduce `NO_DOWNMAP_ERASED_FILE_AST_ID_MARKER`, which prevents `Span`s
from being mapped down into macro expansions.
This is a preparatory step for adding a new field to the
`rust-project.json` format that can inject crate-level attributes.
`Span`s for those attributes will be marked with
`NO_DOWNMAP_ERASED_FILE_AST_ID_MARKER`, indicating that they should not
be mapped down into macro expansions.
rustdoc has a separate environment variable for banning warnings, so
set that in the GitHub action configuration.
https://github.com/rust-lang/cargo/issues/8424#issuecomment-1070988443
Fix all the rustdoc warnings on unknown types or functions. I've
updated references wherever it's obvious, otherwise I've replaced the
rustdoc link with plain backticks.
There were also some cases where rustdoc links referred to private
APIs. I've disabled the rustdoc private API warning in those crates.
We need it because `global_asm!()` is an item. It's unfortunate that such thing can slip in and I see no way to automatically catch that, but thankfully analysis-stats on self has caught that.
Instead of simple numbering, we hash important bits, like the name of the item.
This will allow for much better incrementality, e.g. when you add an item. Currently, this invalidates the IDs of all following items, which invalidates pretty much everything.
And make more queries non-interned.
Also flip the default for queries, now the default is to not intern and to intern a query you need to say `invoke_interned`.
More correctly, *also* per-token. Because as it turns out, while the top-level edition affects parsing (I think), the per-token edition affects escaping of identifiers/keywords.
Because it took me hours to figure out that contrary to common sense, the offset stored is the *end* of the node, and we search by the *start*. Which is why we need a convoluted `partition_point()` instead of a simple `binary_search()`. And this was not documented at all. Which made me make mistakes with my implementation of `SpanMap::merge()`.
The other bug fixed about span map merging is correctly keeping track of the current offset in presence of multiple sibling macro invocations. Unrelated, but because of the previous issue it took me hours to debug, so I figured out I'll put them together for posterity.
