feat: add layout detection (#228)

* feat: create first format modules

* add: example file

* add: structured output formatter

* fix: all parsers outputs list of elements & compatibility formatters

* feat: new basemodel for document

* add: structured output

* fix: test

* fix: add uncategorized text handling

* add: skip on flaky pdf

* add: section block

* fix: change load logic & reate page element

* fix: add pages

* add: split onnxtr det and reco

* feat: Doctr in MegaParse

* fix : Update ReadMe

* fix: add config as constructor parameters

* add: to_numpy to bbox

* add: onnx layout detection

* feat: add category to Document Blocks

* add: List handling and table / image ingestion

* structured output

* blk100 ignore

* fix

* feat: fix output

---------

Co-authored-by: chloedia <[email protected]>

libs/megaparse_sdk/megaparse_sdk/schema/mp_outputs.py

@@ -1,6 +1,6 @@
 from enum import Enum, auto
+from typing import Dict
 
-from llama_parse.base import Dict
 from pydantic import BaseModel, Field
 
 from megaparse_sdk.schema.document import Document