feat(huggingFace): add image task family via ImageTaskCodegen

[PG1204]

⚠️ This PR is stacked on #5278. Until that lands, the diff below also includes #5278's operator + codegen + spec changes. The new code in this PR is codegen/ImageTaskCodegen.scala, the image-related additions to codegen/PythonCodegenBase.scala, the new image fields on HuggingFaceInferenceOpDesc.scala, the frontend image-upload component, and the image-task tests in HuggingFaceInferenceOpDescSpec.scala. Once #5278 merges, this diff will auto-clean to ~856 lines.

What changes were proposed in this PR?

Adds the image task family — 9 HF pipeline tasks — as the second TaskCodegen plugged into the dispatcher established by #5278:

image-only: image-classification, object-detection, image-segmentation, image-to-text
image + prompt: visual-question-answering, document-question-answering, zero-shot-image-classification, image-text-to-text, image-to-image

• codegen/ImageTaskCodegen.scala supplies the per-task payload + parse Python branches for all 9 tasks.
• TaskCodegen trait gains a tasks: Set[String] default method (defaults to Set(task)) so a single codegen can register under multiple task strings; ImageTaskCodegen is the first multi-task codegen to use it.
• CodegenContext extended with imageInput + inputImageColumn (EncodableString).
• HuggingFaceInferenceOpDesc.scala gains 2 new @JsonProperty fields and registers ImageTaskCodegen via the new tasks flat-map.

PythonCodegenBase.scala grows to host the shared image infrastructure:

• Task-family tuples (image_only_tasks, image_prompt_tasks, image_tasks) + image_headers in process_table.
• Per-row image-bytes resolution from upload or column with _read_image_input / _read_binary_value / _compress_image_bytes.
• _post_with_fallback extended with raw_binary_headers + use_raw_binary_body; adds image-text-to-text chat-completions and model-author vision branches.
• _call_provider gains zai-org, Replicate predictions + polling, Fal-ai, Wavespeed submit+poll branches, and image embedding for OpenAI-compatible / unknown-provider fallbacks.
• Image content-type response handling returns data:image/...;base64,... URLs.
• Image helpers added: _read_image_input, _compress_image_bytes, _image_input_as_base64, _read_binary_value, _looks_like_html, _html_to_image_bytes, _extract_json_arg, _url_to_data_url.

Frontend integration (HF lines only — no agent / dataset noise):
HuggingFaceImageUploadComponent declared in app.module.ts, huggingface-image-upload formly type registered, image upload component .ts/.html/.scss + HuggingFace.png + sample-image.png assets.

User-input strings continue to flow through pyb"..." + EncodableString so they reach Python as self.decode_python_template('<base64>') rather than raw literals. PythonCodeRawInvalidTextSpec still passes
(117/117 descriptors py_compile cleanly).

Any related issues, documentation, or discussions?

• Tracking issue: Add image task family (ImageTaskCodegen) to HuggingFace operator #5319
• Closes: Add image task family (ImageTaskCodegen) to HuggingFace operator #5319
• Stacked on: feat(huggingFace): refactor operator into per-task codegen + text-generation #5278 (operator + text-generation — issue Add HuggingFaceInferenceOpDesc with dispatcher + per-task codegen architecture (text-generation) #5277)
• Parent issue: Add Hugging Face inference operator #5041
• Closed sibling issue: Add HuggingFaceModelResource REST endpoints for HF operator UI #5134 (REST resource — landed via feat(huggingFace): add HuggingFaceModelResource for model browsing and media proxy #5124)

How was this PR tested?

• sbt "WorkflowOperator/compile; WorkflowOperator/Test/compile" clean.
• sbt scalafmtCheck clean.
• sbt "WorkflowOperator/testOnly org.apache.texera.amber.operator.huggingFace.HuggingFaceInferenceOpDescSpec" — 18/18 pass (PR 2's 13 spec tests + 5 new image-task tests: image-only routing, VQA / document-QA payload, image-text-to-text chat-completions, image-to-image data-URL parse, all-9-tasks dispatcher coverage).
• sbt "WorkflowOperator/testOnly org.apache.texera.amber.util.PythonCodeRawInvalidTextSpec" — 117/117 descriptors py_compile cleanly with the new operator code paths, no marker leaks.
• Generated Python verified via python3 -m py_compile on sample image-task outputs.

Was this PR authored or co-authored using generative AI tooling?

Yes, co-authored with Claude Opus 4.7.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 60.58394% with 54 lines in your changes missing coverage. Please review.
✅ Project coverage is 53.03%. Comparing base (33903e1) to head (b56f3ac).

Files with missing lines	Patch %	Lines
...mage-upload/hugging-face-image-upload.component.ts	50.00%	41 Missing and 1 partial ⚠️
...ge-upload/hugging-face-image-upload.component.html	38.88%	11 Missing ⚠️
...ber/operator/huggingFace/codegen/TaskCodegen.scala	75.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##               main    #5320    +/-   ##
==========================================
  Coverage     53.02%   53.03%            
  Complexity     2657     2657            
==========================================
  Files          1094     1097     +3     
  Lines         42286    42420   +134     
  Branches       4541     4556    +15     
==========================================
+ Hits          22423    22496    +73     
- Misses        18554    18610    +56     
- Partials       1309     1314     +5     

Flag	Coverage Δ		*Carryforward flag
access-control-service	70.91% <ø> (ø)
agent-service	34.36% <ø> (ø)		Carriedforward from 9ab3e60
amber	53.42% <97.05%> (+0.03%)	⬆️
computing-unit-managing-service	1.65% <ø> (ø)
config-service	56.71% <ø> (ø)
file-service	57.06% <ø> (ø)
frontend	47.87% <48.54%> (+<0.01%)	⬆️
pyamber	89.77% <ø> (ø)		Carriedforward from 9ab3e60
python	90.73% <ø> (ø)		Carriedforward from 9ab3e60
workflow-compiling-service	58.69% <ø> (ø)

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[PG1204]

/request-review @Ma77Ball

[Ma77Ball]

Please look at the suggestions below.

[github-actions]

✅ No material benchmark regressions detected

🟢 6 better · 🔴 0 worse · ⚪ 9 noise (<±5%) · 0 without baseline

Compared against main 33903e1 benchmarked on this same runner, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated.

Dashboard · Run

	config	throughput	MB/s	latency	max Δ latest / 7d
🟢	bs=10 sw=10 sl=64	437	0.267	21,470/32,276/32,276 us	🟢 -5.3% / 🟢 -9.7%
🟢	bs=100 sw=10 sl=64	953	0.581	106,635/120,610/120,610 us	🟢 -23.1% / 🟢 -13.7%
🟢	bs=1000 sw=10 sl=64	1,119	0.683	901,582/922,952/922,952 us	🟢 -7.4% / 🟢 -9.8%

Baseline details

Latest main 33903e1 from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	437 tuples/sec	442 tuples/sec	410.82 tuples/sec	-1.1%	+6.4%
bs=10 sw=10 sl=64	MB/s	0.267 MB/s	0.27 MB/s	0.251 MB/s	-1.1%	+6.5%
bs=10 sw=10 sl=64	p50	21,470 us	20,615 us	23,785 us	+4.1%	-9.7%
bs=10 sw=10 sl=64	p95	32,276 us	34,099 us	34,980 us	-5.3%	-7.7%
bs=10 sw=10 sl=64	p99	32,276 us	34,099 us	34,980 us	-5.3%	-7.7%
bs=100 sw=10 sl=64	throughput	953 tuples/sec	930 tuples/sec	891.94 tuples/sec	+2.5%	+6.8%
bs=100 sw=10 sl=64	MB/s	0.581 MB/s	0.567 MB/s	0.544 MB/s	+2.5%	+6.7%
bs=100 sw=10 sl=64	p50	106,635 us	104,082 us	112,277 us	+2.5%	-5.0%
bs=100 sw=10 sl=64	p95	120,610 us	156,890 us	139,802 us	-23.1%	-13.7%
bs=100 sw=10 sl=64	p99	120,610 us	156,890 us	139,802 us	-23.1%	-13.7%
bs=1000 sw=10 sl=64	throughput	1,119 tuples/sec	1,107 tuples/sec	1,041 tuples/sec	+1.1%	+7.5%
bs=1000 sw=10 sl=64	MB/s	0.683 MB/s	0.676 MB/s	0.635 MB/s	+1.0%	+7.5%
bs=1000 sw=10 sl=64	p50	901,582 us	893,996 us	972,714 us	+0.8%	-7.3%
bs=1000 sw=10 sl=64	p95	922,952 us	996,521 us	1,023,057 us	-7.4%	-9.8%
bs=1000 sw=10 sl=64	p99	922,952 us	996,521 us	1,023,057 us	-7.4%	-9.8%

Raw CSV

config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,457.24,200,128000,437,0.267,21470.16,32275.64,32275.64
1,100,10,64,20,2099.62,2000,1280000,953,0.581,106634.63,120609.75,120609.75
2,1000,10,64,20,17874.12,20000,12800000,1119,0.683,901581.83,922951.99,922951.99