[ hnrm2 ] Use precision-enhanced hnrm2

[skykongkong8]

Along with #2555 I inspected through every custom-made half-precision calculation function, and hnrm2 is the only one left that needs f16-f32 precision supplement.

Proposed Changes

• Previous hnrm2 was using full-fp16.
• Since this is also one of dimension-shrinking computation, should use inter-fp32 values to enhance precision.
• This has not been detected due to small dimension Tensor usage in unittest. Add higher dimension test case accordingly.
• Note that this function is responsible for Tensor::l2norm(), frequently used for mse loss computation.

Differnece w.r.t. fp32-cblas

dim	f16 (prev)	f16f32 (now)
768	282.233	0.232574

Latency

mean value ( TC = 100) Since this is nanosec-unit result, almost trivial

dim	f16	f16f32	fp32-cblas
768	508 ns	579 ns	4242 ns

Better accuracy with almost no latency deterioration

Self evaluation:

1. Build test: [X]Passed [ ]Failed [ ]Skipped
2. Run test: [X]Passed [ ]Failed [ ]Skipped

[taos-ci]

:memo: TAOS-CI Version: 1.5.20200925. Thank you for submitting PR #2556. Please a submit 1commit/1PR (one commit per one PR) policy to get comments quickly from reviewers. Your PR must pass all verificiation processes of cibot before starting a review process from reviewers. If you are new member to join this project, please read manuals in documentation folder and wiki page. In order to monitor a progress status of your PR in more detail, visit http://ci.nnstreamer.ai/.