Replace most direct calls to `TRawInstruction()` with calls to `getInstructionTranslatedElement()` and `getInstructionTag()`, matching existing practice. One tiny RA diff in an inconsequential join order in `getInstructionVariable`.
This noopt predicate is no longer necessary. It's equivalent to `instruction = TRawInstruction(element, tag)`, which is already materialized and has a more favorable column order anyway.
These predicates are only used within the new single IR stage, so there's no need to cache them beyond that. RA diffs are trivial. Where previously many of the predicate on `Instruction` were inline wrappers around cached predicates from `IRConstruction`, now the predicates from `IRConstruction` get inlined into the `Instruction` predicates, and the `Instruction` predicates get materialized. The net amount of work is the same, but now it's not getting cached unnecessarily.
I think these queries have excellent results on lgtm.com. Many of the
results come from projects that use `sprintf` like it's a templating
engine, trusting that values from `argv` or `getenv` contain the correct
number of `%s`. I think we want to flag that.
The structure of the change note is modeled after 91af51cf46.
Some of our IR consistency failure query predicates already produced results in the schema as an `@kind problem` query, including `$@` replacements for the enclosing `IRFunction` to make it easier to figure out which function to dump when debugging. This change moves the rest of the query predicates in `IRConsistency.qll` to do the same. In addition, it wraps each call to `getEnclosingIRFunction()` to return an `OptionalIRFunction`, which can be either a real `IRFunction` or a placeholder in case `getEnclosingIRFunction()` returned no results. This exposes a couple new consistency failures in `syntax-zoo`, which will be fixed in a subsequent commit.
This change also deals with consistency failures when the enclosing `IRFunction` has more than one `Function` or `Location`. For multiple `Function`s, we concatenate the function names. For multiple `Location`s, we pick the first one in lexicographical order. This changes the number of results produced in the existing tests, but does't change the actual number of problems.
