Enable models regressed because of frontend type alignment changes (openvinotoolkit#26661)

Enable models regressed by frontend type alignment changes

Enable models regressed by frontend type alignment changes aeaa4f7 &
e732ade that insert additional ConvertLike operations that prevent
SDPAToPagedAttention transformation to work smoothly as expected.

Fix it by moving ConvertConvertLike after type resolution in
FrontEnd::normalize() to turn ConvertLike into Convert only after types
are resolved hence nothing prevents SDPAToPagedAttention transformation
to do its job.

Therefore, enable katuni4ka/tiny-random-chatglm2 &
katuni4ka/tiny-random-glm4 in precommit tests.

List of enabled models:
* THUDM/chatglm2-6b
* THUDM/chatglm3-6b
* katuni4ka/tiny-random-chatglm2
* katuni4ka/tiny-random-glm4

Ticket:
	* CVS-152286

Author: CuriousPanCake

src/frontends/pytorch/src/frontend.cpp

@@ -317,6 +317,7 @@ void FrontEnd::normalize(const std::shared_ptr<ov::Model>& model) const {
     manager.register_pass<ov::pass::RemoveMultiSubGraphOpDanglingParamsResults>();
     manager.register_pass<ov::pass::ReverseShapeAndTypeInfer>();
     manager.register_pass<ov::pass::ResolveNameCollisions>(true);
+    manager.register_pass<ov::pass::ConvertConvertLike>();
     manager.run_passes(model);
 
     // Usually if nn.Module.forward is given as a source model for conversion, there is the first Parameter

tests/model_hub_tests/transformation_tests/models/hf-tiny-random-models-precommit

@@ -39,8 +39,8 @@ fxmarty/really-tiny-falcon-testing,https://huggingface.co/fxmarty/really-tiny-fa
 Xenova/tiny-random-Phi3ForCausalLM,https://huggingface.co/Xenova/tiny-random-Phi3ForCausalLM
 facebook/opt-125m,https://huggingface.co/facebook/opt-125m
 facebook/opt-350m,https://huggingface.co/facebook/opt-350m
+katuni4ka/tiny-random-chatglm2,https://huggingface.co/katuni4ka/tiny-random-chatglm2
+katuni4ka/tiny-random-glm4,https://huggingface.co/katuni4ka/tiny-random-glm4
 hf-internal-testing/tiny-random-BioGptForCausalLM,https://huggingface.co/hf-internal-testing/tiny-random-BioGptForCausalLM,xfail,No ScaledDotProductAttention operation observed in the graph CVS-145820
 hf-internal-testing/tiny-random-XGLMForCausalLM,https://huggingface.co/hf-tiny-model-private/tiny-random-XGLMForCausalLM,xfail,No ScaledDotProductAttention operation observed in the graph CVS-145820
-katuni4ka/tiny-random-orion,https://huggingface.co/katuni4ka/tiny-random-orion,xfail,No ScaledDotProductAttention operation observed in the graph CVS-145820
-katuni4ka/tiny-random-chatglm2,https://huggingface.co/katuni4ka/tiny-random-chatglm2,xfail,Model references undeclared parameters: beam_idx () CVS-145820
-katuni4ka/tiny-random-glm4,https://huggingface.co/katuni4ka/tiny-random-glm4,xfail,Model references undeclared parameters beam_idx () attention_mask () CVS-145820
+katuni4ka/tiny-random-orion,https://huggingface.co/katuni4ka/tiny-random-orion,xfail,No ScaledDotProductAttention operation observed in the graph CVS-145820