* Add additional function for parsing traversals with [*] keys

* add more context around skipped test cases

Co-authored-by: hashicorp-tsccr[bot] <hashicorp-tsccr[bot]@users.noreply.github.com>

We were calling .Range() on any unknown sourceVal, without first checking
whether it was marked. That method panics if called on a marked value,
so we need to strip that off first.

While testing this I found some return paths that weren't properly
transferring the source value's marks to the output, and so this also
addresses those so that all return paths preserve whatever markings are
present on the source value.

In particular, if a non-list/set/tuple value gets "upgraded" into a tuple
then we must transfer its marks onto the tuple, because the decision about
constructing that value was based on characteristics of the source value.

Previously if the for_each expression was marked then expansion would
fail because marked expressions are never directly iterable.

Now instead we'll allow marked for_each and preserve the marks into the
values produced by the resulting block as much as we can. This runs into
the classic problem that HCL blocks are not values themselves and so
cannot carry marks directly, but we can at least make sure that the values
of any leaf arguments end up marked.

Similar to the previously-added UnknownBody, the new optional interface
MarkedBody allows hcl.Body implementations to suggest a set of marks that
ought to be applied to any value that's generated to represent the content
of that body.

The dynblock extension then uses this to get hcldec to mark the whole
object representing any block that was generated by a dynamic block whose
for_each was marked, for a better representation of the fact that a
block's existence was decided based on a marked value.

Co-authored-by: hashicorp-tsccr[bot] <hashicorp-tsccr[bot]@users.noreply.github.com>