Prior to this commit, a build that used multiple targets needed to
provide `tvm::build` with a `Map<Target, IRModule>` specifying which
target should be used to compile each `IRModule`.  As a result,
lowering passes could not introduce new targets based on a PrimFunc's
content (e.g. a `with T.target()` frame to delegate out to another
device), nor simplify based on cross-device subroutines (e.g. simplify
a host-side conditional based on the known output of a device-side
internal subroutine).

This commit makes the `tvm::attr::kTarget` attribute (`"target"`) be
the single source of truth for where a `PrimFunc` will be executed.
Other existing methods for specifying the target (the `target`
parameter for `tvm.build`, the keys in a `Map<Target,IRModule>`, the
parameter to the pass `tir::transform::BindTarget`) are still accepted
as inputs, and may provide a default value for `tvm::attr::kTarget` if
the attribute is missing, but may not overwrite the target attribute.

This is part of a series of commits to simplify the handling of
multi-target builds.

If a flattened buffer is produced for use in `BufferLoad` and
`BufferStore` statements, generate a `DeclBuffer`.

This is a subset of the changes made in
apache#14778, broken out for ease of
testing and review.

When producing a flattened buffer for use in `BufferLoad` and
`BufferStore` nodes, generate a `DeclBuffer` for the flattened buffer.

This is a subset of the changes made in
apache#14778, broken out for ease of
testing and review.