further tweak tensor and block show

[Jutho]

@lkdvos , this tries to get some sensible default for showing some blocks while respecting the displaysize when :limit = true. Since standard repl printing sets :limit = true, this is the default printing.

However, with the current implementation, this also means that the blocks will be truncated (both the matrices itself and the number of blocks shown) in both the output of the tensor t and in the output of blocks(t).

There might be a case to be made that, if a user calls blocks(t) explicitly as a REPL prompt, the user wants to see all blocks, and so we should "ignore" the default :limit = true setting of REPL printing, at least for truncating the number of blocks, but we could still pass on :limit = true to the individual block printing. This would then be analogous to how an array of size n x n x 10 with large n is shown. All 10 slices along the 3rd dimension are shown, but every slice is truncated such that single slice fits the display.

[github-actions]

Your PR no longer requires formatting changes. Thank you for your contribution!

[codecov]

Codecov Report

❌ Patch coverage is 91.89189% with 3 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/tensors/blockiterator.jl	89.65%	3 Missing ⚠️

Files with missing lines	Coverage Δ
src/tensors/abstracttensor.jl	57.07% <100.00%> (+5.11%)	⬆️
src/tensors/blockiterator.jl	38.61% <89.65%> (+23.79%)	⬆️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[lkdvos]

There might be a case to be made that, if a user calls blocks(t) explicitly as a REPL prompt, the user wants to see all blocks, and so we should "ignore" the default :limit = true setting of REPL printing, at least for truncating the number of blocks, but we could still pass on :limit = true to the individual block printing. This would then be analogous to how an array of size n x n x 10 with large n is shown. All 10 slices along the 3rd dimension are shown, but every slice is truncated such that single slice fits the display.

I think I agree with this, which might mean that we really just want a different show method for blocks vs for TensorMap?

Otherwise this definitely looks like a great improvement I can get behind!

[Jutho]

Ok, I've changed the implementation such that explicitly calling blocks(t) will not truncate all the blocks being shown (but limit is still in effect when showing the individual blocks)

[lkdvos]

Do you think it would make sense to have some dummy tests that just checks that the output of showing a tensor is what it is now, such that we get notified if we (accidentally/intentionally) change this in the future?

Otherwise this looks ready for me.

[lkdvos]

I merged the other PR in hope of fixing the CI here, which would then automerge this tonight.