Feat/vml grammar

[joswig]

This PR encompasses initial support for VML 2.0.11 editing. The grammar is based on DOC-000477 Rev C., with minor changes related to Bison vs Lezer differences, and some minor implementation choices when grammar appeared to have a typo.

VML capabilities are mapped to file extension.

Sample sequences and dictionaries are available from @joswig, but are not cleared for release, so treat as ITAR.

Follow up actions

• More polish on context awareness on content assist, there's a minimal implementation to show potential capabilities. Low hanging fruit to add more, but trying to get this PR merged.
• Add support for block library, once that's done then treat spawn and call statements in similar manner to issue
• Consider moving CDL dictionary parsing code to the server. We should decide on XTCE path forward first, but the current implementation is inconsistent with rest of application
• more testing, only 2 test sequences have been provided, will work with missions as part of User Acceptance Testing

[duranb]

I'm a little concerned with the vmlMode for SequenceEditor because it bypasses the adaptation flow. Can we do a follow up right after this is merged? I know there's a chance that this will be pulled out of core, but I just want to make sure that in the off chance that it does stay in, that it's integrated properly to work in conjunction with adaptations. For example, right now adaptations require an inputFormat. So if this is baked into Aerie, we should handle cases where they don't match or make the inputFormat drive vmlMode instead of the sequence name.

[cohansen]

Did a cursory glance and it all looked good. Still curious how our decisions with language support will effect this.

[joswig]

I'm a little concerned with the vmlMode for SequenceEditor because it bypasses the adaptation flow. Can we do a follow up right after this is merged? I know there's a chance that this will be pulled out of core, but I just want to make sure that in the off chance that it does stay in, that it's integrated properly to work in conjunction with adaptations. For example, right now adaptations require an inputFormat. So if this is baked into Aerie, we should handle cases where they don't match or make the inputFormat drive vmlMode instead of the sequence name.

That sounds good, I think it will be a couple of steps to get there and publishing the adaptation types is one of the next ones up.

[duranb]

lgtm!

[joswig]

Also will follow up with

#1534