[Fix][Tutorial][VTA] Update tune_relay_vta.py to support single board

[insop]

• support single pynq board run, change is credited to @i24361, https://github.com/i24361/incubator-tvm/blob/0472b1f347976229a29be8a6e60b626a0604c8df/vta/tutorials/autotvm/tune_relay_vta_with_one_board.py
• fixes the save fail
• issues and changes are discussed in https://discuss.tvm.apache.org/t/vta-workaround-for-autotuning-with-one-pynq-z1-board/8091/9

Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from Reviewers by @ them in the pull request thread.

[insop]

With this change on top of the latest TVM (28647f2), here is an example log with a single board.

$ python vta/tutorials/autotvm/tune_relay_vta.py
Extract tasks...
...100%, 0.06 MB, 174 KB/s, 0 seconds passed
...100%, 0.73 MB, 1274 KB/s, 0 seconds passed
Extracted 10 conv2d tasks:
(1, 14, 14, 256, 512, 1, 1, 0, 0, 2, 2)
(1, 28, 28, 128, 256, 1, 1, 0, 0, 2, 2)
(1, 56, 56, 64, 128, 1, 1, 0, 0, 2, 2)
(1, 56, 56, 64, 64, 3, 3, 1, 1, 1, 1)
(1, 28, 28, 128, 128, 3, 3, 1, 1, 1, 1)
(1, 56, 56, 64, 128, 3, 3, 1, 1, 2, 2)
(1, 14, 14, 256, 256, 3, 3, 1, 1, 1, 1)
(1, 28, 28, 128, 256, 3, 3, 1, 1, 2, 2)
(1, 7, 7, 512, 512, 3, 3, 1, 1, 1, 1)
(1, 14, 14, 256, 512, 3, 3, 1, 1, 2, 2)
Tuning...
[Task  1/10]  Current/Best:   17.63/ 116.35 GFLOPS | Progress: (480/480) | 255.47 s Done.
[Task  2/10]  Current/Best:   19.43/ 137.07 GFLOPS | Progress: (576/576) | 291.50 s Done.
[Task  3/10]  Current/Best:    0.00/ 137.87 GFLOPS | Progress: (672/1000) | 365.49 s

[insop]

@tmoreau89

[insop]

I found the change that I picked was already in the PR by created a separate file.

• [TUTORIAL][FIX] Fix VTA autotuning from tutorial fails with one P… #6688

[insop]

@i24361 , @merrymercy

[tmoreau89]

Thank you for the fixes @insop and bringing this script up to date!

[tmoreau89]

The PR has been merged.