feat: make local context available in data-items loop

[jfmcquade]

PR Checklist

• PR title descriptive (can be used in release notes)

TODO

• Builds on test: fix issue with e2e-data parse script #2305, merge that first for clearer diffs
• Builds on the spec test files added by feat: set_item action: optionally target an index or id #2215, although should be able to be merged independently and resolve any conflicts afterwards

Description

Allows rows within a data-items loop to reference local variables (#2303).

Dev notes

This feels like a hack, and adding another caveat to the template-variables service logic doesn't feel great, but it does seem to work.

NB: view the code changes for template-variables.service.ts with whitespace changes hidden in order to see meaningful diff.

Git Issues

Closes #2303

Screenshots/Videos

debug_data_items

debug_data_items_local

[jfmcquade]

Because calcContext gets overridden by templateCalcService.getCalcContext() as part of evaluatePLHString(), the values here has no effect on the results of evaluatePLHData() in current test cases

[esmeetewinkel]

I found this works in general, but there is a case where somehow the first iteration of a local variable in a data-items loop is found, but next iterations are not. I thought it had to do with the presence of a display group, but that doesn't seem to be the case since this template works.

debug_kids

[chrismclarke]

I'm slightly reluctant to add more case handling to the template-variable service as (like you mentioned), it's already feeling pretty difficult to follow/maintain.

I've proposed an alternate fix in #2314, let me know what you think and whether there would still be use to migrate any of the tests/notes from this PR over