M6: PDF export depth and REST conditional and variable-mode actions (#65)

PDF export
- Vector masks clip to the mask's outline, luminance masks are luminosity
  soft masks (alpha masks as before).
- Background blur: the backdrop the page draws under the layer is rendered,
  blurred and cut to its shape, under the layer's vector content.
- Text hyperlinks become clickable link annotations (one per line and link).
- Frames-to-PDF shares one path (`frames_pdf`): sRGB conversion for Display
  P3 files, hidden frames left out, the file name as the PDF title.
- `figcraft-cli export -o x.pdf` and `convert x.figcraft x.pdf` write the
  page's frames one PDF page each.
- Tests: mask types, backdrops, links, image deduplication, page sizes in
  points, hostile inputs; CLI end to end. Checked by rendering the PDFs back
  to PNG and diffing against the PNG export.

Prototyping
- Model: conditional actions (if / else if / else blocks), set variable mode,
  and expressions (literals, variable references, + - * /, comparisons,
  and/or/not, negate) for set-variable values and conditions; walk helpers
  for nested actions (remap on paste, colours, destination checks, library
  copies and pruning, canvas connections).
- REST import maps CONDITIONAL, SET_VARIABLE_MODE and SET_VARIABLE with
  variable references and EXPRESSION values; unsupported functions (mode
  lookups) skip the action with a warning.
- The presenter runs them in order (`figcraft_engine::proto_run`); the
  Prototype panel summarises conditionals, edits set-variable-mode, and
  `ui.set {protoEditor}` opens an interaction's editor.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

docs/cli.md

@@ -34,18 +34,23 @@ page 1 "Screens": 380 nodes, 4 top-level
 - `--node` takes a layer id (`12` or `0:12`) or a layer name.
 - The format comes from `--format`, else `<out>`'s extension, else png. `--scale` (default 1, max 64)
   applies to raster formats.
+- A `.pdf` `<out>` without `--node` holds the page's top-level frames, one PDF page each in reading
+  order, each the frame's size in points (a page without frames goes on one page). Shapes, text
+  and gradients stay vector, text is real (selectable, with subset fonts) and its links clickable;
+  shadows and blurs are images.
 
 ```sh
 figcraft-cli export app.figcraft -o exports/ --scale 2
 figcraft-cli export app.figcraft -o hero.svg --node "Hero"
+figcraft-cli export app.figcraft -o screens.pdf --page "Screens"
 ```
 
 ### `convert <in> <out>`
 
 By extension. Inputs: `.figcraft`, `.json` (FigCraft or REST API file JSON), `.svg` (becomes vectors), `.png/.jpg/.webp/.gif/.bmp`
 (an image layer). Outputs: `.figcraft`, `.json` (FigCraft JSON, pretty-printed), `.svg`,
-`.png/.jpg/.webp` (the current page rendered at 1x with its background), `.pdf` (when PDF export is
-available).
+`.png/.jpg/.webp` (the current page rendered at 1x with its background), `.pdf` (the current page's
+frames, one page each).
 
 ### `render <file> -o <out.png> [--scale N] [--page NAME]`