[CLI]: CLI Crash Guards (#15477)

# Description

<!--
> Please provide a summary of the changes made in this PR. Include
details such as:
  > * What issue does this PR address or fix?
  > * What new features or enhancements does this PR introduce?
> * Are there any breaking changes or dependencies that need to be
considered?
-->

Part 1 of 3 of the CLI-mode bug sweep, split out of #15452 per review
feedback there. This PR contains the crash fixes.

## Fixes

- **`--outputdir`/`--datadir` with a missing parent directory aborted**
via unguarded `create_directory`. Directories are now created
recursively, with a graceful early exit and a specific error message if
creation fails.
- **`--slice` + `--export-3mf` segfaulted on a from-scratch slice**:
`ConfigOptionVector::get_at()` on an empty vector is `.front()` of an
empty vector (UB). Guards added for `filament_color`/`filament_id` at
the CLI call site, and inside `DynamicPrintConfig::get_filament_type`
for `filament_type`/`filament_is_support`/`filament_id`. Only *empty*
vectors are treated as missing — the existing clamp-to-front behavior
for merely out-of-range indices is preserved, so GUI callers are
unaffected.
- **OOB heap write from stale `filament_self_index` on
`--load-filaments`** (fixes #14181): a 3MF carrying more
`filament_self_index` entries than loaded filaments wrote past the end
of `old_variant_counts`. The guard validates both bounds — entries `>
filament_count` *and* non-positive entries (`< 1`), since a single `0`
in an otherwise-valid array indexes `old_variant_counts[-1]`.
- **Wrong printable-area check** for non-rectangular beds: use the
printable area's bounding box instead of a naive vertex calculation
(fixes #15363).
- **`nozzle_height` and `align_center` were not read into the arrange
config** in CLI mode.

# Screenshots/Recordings/Graphs

<!--
> Please attach relevant screenshots to showcase the UI changes.
> Please attach images that can help explain the changes.
-->

## Tests

<!--
> Please describe the tests that you have conducted to verify the
changes made in this PR.
-->
- Repro'd each crash on CLI before the fix; all resolved after.
- `tests/libslic3r` suite passes; full binary builds clean on Linux.
- Added `get_filament_type` unit tests 

<!--
> A guide for users on how to download the artifacts from this PR.
-->

[How to Download Pull Requests Artifacts for
Testing](https://www.orcaslicer.com/wiki/how_to_download_pr_artifacts)

src/libslic3r/PrintConfig.cpp

@@ -9865,7 +9865,15 @@ std::string DynamicPrintConfig::get_filament_type(std::string &displayed_filamen
     auto* filament_type = dynamic_cast<const ConfigOptionStrings*>(this->option("filament_type"));
     auto* filament_is_support = dynamic_cast<const ConfigOptionBools*>(this->option("filament_is_support"));
 
-    if (!filament_type)
+    // get_at() on an empty vector option is undefined behavior (.front() of an empty vector),
+    // and e.g. filament_id is never populated on a CLI from-scratch slice - treat an empty
+    // option the same as a missing one.
+    if (filament_id && filament_id->values.empty())
+        filament_id = nullptr;
+    if (filament_is_support && filament_is_support->values.empty())
+        filament_is_support = nullptr;
+
+    if (!filament_type || filament_type->values.empty())
         return "";
 
     if (!filament_is_support) {