Add Syriac script support (#13800)

* Add Syriac Language support dictionary

The Syriac Script is a Unicode block containing characters for all forms of the Syriac alphabet, including the Estrangela, Serto, Eastern Syriac, and the Christian Palestinian Aramaic variants. It is used in Literary Syriac, Neo-Aramaic, and Arabic among Syriac-speaking Christians. It was used historically to write Armenian, Persian, Ottoman Turkish, and Malayalam. The script, like Arabic and Hebrew is RTL.

https://en.wikipedia.org/wiki/Syriac_(Unicode_block)
https://en.wikipedia.org/wiki/Syriac_language

* Add Syriac script support for training

The Syriac Script is a Unicode block containing characters for all forms of the Syriac alphabet, including the Estrangela, Serto, Eastern Syriac, and the Christian Palestinian Aramaic variants. It is used in Literary Syriac, Neo-Aramaic, and Arabic among Syriac-speaking Christians. It was used historically to write Armenian, Persian, Ottoman Turkish, and Malayalam. The script, like Arabic and Hebrew is RTL.

https://en.wikipedia.org/wiki/Syriac_(Unicode_block)
https://en.wikipedia.org/wiki/Syriac_language

configs/rec/multi_language/rec_syriac_lite_train.yml

@@ -0,0 +1,110 @@
+Global:
+  use_gpu: true
+  epoch_num: 500
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec_syriac_lite
+  save_epoch_step: 3
+  eval_batch_step:
+  - 0
+  - 2000
+  cal_metric_during_train: true
+  pretrained_model: null
+  checkpoints: null
+  save_inference_dir: null
+  use_visualdl: false
+  infer_img: null
+  character_dict_path: ppocr/utils/dict/syriac_dict.txt
+  max_text_length: 25
+  infer_mode: false
+  use_space_char: true
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+  regularizer:
+    name: L2
+    factor: 1.0e-05
+Architecture:
+  model_type: rec
+  algorithm: CRNN
+  Transform: null
+  Backbone:
+    name: MobileNetV3
+    scale: 0.5
+    model_name: small
+    small_stride:
+    - 1
+    - 2
+    - 2
+    - 2
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn
+    hidden_size: 48
+  Head:
+    name: CTCHead
+    fc_decay: 1.0e-05
+Loss:
+  name: CTCLoss
+PostProcess:
+  name: CTCLabelDecode
+Metric:
+  name: RecMetric
+  main_indicator: acc
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: train_data/
+    label_file_list:
+    - train_data/syriac_train.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - RecAug: null
+    - CTCLabelEncode: null
+    - RecResizeImg:
+        image_shape:
+        - 3
+        - 32
+        - 320
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: true
+    batch_size_per_card: 256
+    drop_last: true
+    num_workers: 8
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: train_data/
+    label_file_list:
+    - train_data/syriac_val.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - CTCLabelEncode: null
+    - RecResizeImg:
+        image_shape:
+        - 3
+        - 32
+        - 320
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: false
+    drop_last: false
+    batch_size_per_card: 256
+    num_workers: 8

ppocr/utils/dict/syriac_dict.txt

@@ -0,0 +1,157 @@
+!
+#
+$
+%
+&
+'
+(
++
+,
+-
+.
+/
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+:
+?
+@
+A
+B
+C
+D
+E
+F
+G
+H
+I
+J
+K
+L
+M
+N
+O
+P
+Q
+R
+S
+T
+U
+V
+W
+X
+Y
+Z
+_
+a
+b
+c
+d
+e
+f
+g
+h
+i
+j
+k
+l
+m
+n
+o
+p
+q
+r
+s
+t
+u
+v
+w
+x
+y
+z
+É
+é
+܀
+܁
+܂
+܃
+܄
+܅
+܆
+܇
+܈
+܉
+܊
+܋
+܌
+܍
+܏
+ܐ
+ܑ
+ܒ
+ܓ
+ܔ
+ܕ
+ܖ
+ܗ
+ܘ
+ܙ
+ܚ
+ܛ
+ܜ
+ܝ
+ܞ
+ܟ
+ܠ
+ܡ
+ܢ
+ܣ
+ܤ
+ܥ
+ܦ
+ܧ
+ܨ
+ܩ
+ܪ
+ܫ
+ܬ
+ܭ
+ܮ
+ܯ
+ܰ
+ܱ
+ܲ
+ܳ
+ܴ
+ܵ
+ܶ
+ܷ
+ܸ
+ܹ
+ܺ
+ܻ
+ܼ
+ܽ
+ܾ
+ܿ
+݀
+݁
+݂
+݃
+݄
+݅
+݆
+݇
+݈
+݉
+݊
+ݍ
+ݎ
+ݏ