[Feature] Support LLaVA

[mzr1996]

Support LLavA multimodal algorithm

[codecov]

Codecov Report

Patch coverage: 34.23% and project coverage change: -2.20 ⚠️

Comparison is base (f9dcae2) 68.16% compared to head (709512b) 65.97%.

❗ Current head 709512b differs from pull request most recent head 28f03ac. Consider uploading reports for the commit 28f03ac to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##              dev    #1652      +/-   ##
==========================================
- Coverage   68.16%   65.97%   -2.20%     
==========================================
  Files         295      316      +21     
  Lines       23372    24991    +1619     
  Branches     3713     3979     +266     
==========================================
+ Hits        15932    16488     +556     
- Misses       6880     7900    +1020     
- Partials      560      603      +43     

Flag	Coverage Δ
unittests	65.97% <34.23%> (-2.20%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmpretrain/apis/image_retrieval.py	21.42% <ø> (ø)
mmpretrain/datasets/__init__.py	63.41% <0.00%> (-10.88%)	⬇️
mmpretrain/datasets/gqa_dataset.py	0.00% <0.00%> (ø)
mmpretrain/datasets/nocaps.py	0.00% <0.00%> (ø)
mmpretrain/datasets/ocr_vqa.py	0.00% <0.00%> (ø)
mmpretrain/datasets/scienceqa.py	0.00% <0.00%> (ø)
mmpretrain/datasets/textvqa.py	0.00% <0.00%> (ø)
mmpretrain/datasets/transforms/__init__.py	100.00% <ø> (ø)
mmpretrain/datasets/vizwiz.py	0.00% <0.00%> (ø)
mmpretrain/datasets/vsr.py	0.00% <0.00%> (ø)
... and 39 more

... and 1 file with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.