Fixed to the phase-field model

[shkodm]

Synchronising various fixes I did to the phase-field mode over the past months:

• Split initialisation of wall normals in 2 parts, so it is possible to supply user defined wall normals using RunR
• Fix various edge cases and bugs with wetting boundary conditions
• Fix pressure boundary conditions
• Add markers for the saturation

Pressure boundary conditions fixes:

• Now use proper ZouHe boundary conditions, with 3 variants. (Some can be removed later)

  • fflux enforces constant flux -> Breaks when interface enters the outlet - default (i.e with WPressure or EPRessure name)
  • fixed phase-field (e.g. EPressurefpf) makes sure the inlet / outlet phase-field is the same as specified (i.e. sum of distributions forced to be equal to it). Effect: interface propagation stop at the outlet, "impossible-to-displace liquid". But seems to not work with the inlet
  • open suffix (i.e. EPressureopen) does not enforce the phase-field. Interface propagates into the outlet, splits in 2, there are so strange effects, but otherwise does not look too crazy
  • Gradient / Laplacian on the velocity / pressure boundaries is always zero (does not touch the other side of the domain)
  • Wall normal calculations do not take into account the other side of the domain (periodicity is switched off). This works only for the case of X direction and when the flag InvasionDrainage is non-zero. And when the boundary conditions are the first / last layers of the domain. Otherwise, one has to compile the model with autosym option, and mark the inlet / oulets with Symmetry boundaries, so that the periodicity is switched off
  • InvasionDrainage marks if the case is drainage (non-wetting displacing wetting), or invasion (wetting displacing non-wetting). This sets some densities and phase-field on the inlet / outlet
  • Apply boundary conditions before VTK export, so the inlet / outlet zones don't look strange due to streaming

  Merging with master, because I (and I think other people who are using phase-field model are mostly using master branch and the fixes are quite crucial)

Now doing some testing to make sure I did not mess up when refactoring. So draft PR for now. cc @TravisMitchell

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 36.36%. Comparing base (2362a9d) to head (98accf4). Report is 24 commits behind head on master.

:exclamation: Current head 98accf4 differs from pull request most recent head 9a0f8d6

Please upload reports for the commit 9a0f8d6 to get more accurate results.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #515 +/- ## ========================================== - Coverage 44.24% 36.36% -7.89% ========================================== Files 166 159 -7 Lines 7567 6790 -777 ========================================== - Hits 3348 2469 -879 - Misses 4219 4321 +102 ``` | [Flag](https://app.codecov.io/gh/CFD-GO/TCLB/pull/515/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=CFD-GO) | Coverage Δ | | |---|---|---| | [d2q9](https://app.codecov.io/gh/CFD-GO/TCLB/pull/515/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=CFD-GO) | `?` | | | [d3q27_PSM_NEBB](https://app.codecov.io/gh/CFD-GO/TCLB/pull/515/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=CFD-GO) | `36.36% <ø> (+0.01%)` | :arrow_up: | | [d3q27_pf_velocity](https://app.codecov.io/gh/CFD-GO/TCLB/pull/515/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=CFD-GO) | `?` | | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=CFD-GO#carryforward-flags-in-the-pull-request-comment) to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[shkodm]

@nparaskevas you can test it now if you want. I provided the description of different boundary conditions variants in the PR description. Just using WPressure on the inlet and EPressurefflux or EPressurefpf or EPressureopen on the outlet would work, but you should play around.

Don't forget to add InvasionDrainage value="2" to the setting in case of drainage, or value = 1 in case of invasion